QUT Digital Repository:

http://eprints.qut.edu.au/

Furneaux, Craig W. (2008) Engineering Asset Procurement: A Complex Adaptive

System Perspective. In Brown, Kerry A. and Mandell, Myrna and Furneaux, Craig
W. and Beach, Sandra, Eds. Proceedings Contemporary Issues in Public
Management: The Twelfth Annual Conference of the International Research
Society for Public Management (IRSPM XII), pages pp. 1-34, Brisbane, Australia.

Copyright 2008 (please consult author)

Engineering asset procurement: A complex adaptive system perspective

Author: Craig Furneaux (c.furneaux@qut.edu.au)
Paper presented at the International Research Society for Public Management (IRSPM XII) Doctoral Panel
Please do not cite this paper without permission from the author.

Introduction
The Australian Procurement and Construction Council (APCC) (2001) noted that Australian
government assets, such as buildings, roads, rail and utilities, are valued at around $371billion and
approximately $18 billion is spent annually by state, territory and commonwealth governments on
acquiring and maintaining these assets. Ferguson (2007) estimates that over $200 billion will be
spent on engineering assets such as transport (roads and rail), ports, utilities (water and electricity)
and broadband in the next few years in Australia. Such engineering assets are highly complex
arrangements which comprise social and technical systems, are capital intensive, and typically last
for significant lengths of time (Herder and Verwater-Lukszo 2006). Indeed it is considered that the
optimal functioning of engineering assets such as transportation, energy, information and
communication, and water is vital for the economy and society (Herder and Verwater-Lukszo
2006, 119). Engineering assets are thus significant in both economic and social terms (APCC
2001).
Recognising the importance of engineering assets to society, many jurisdictions in Australia have
developed policies on the strategic management of engineering assets (e.g. Queensland Treasury
2003, Western Australia Department of Treasury and Finance 2005) particularly in order to guide
the procurement of assets, asset management and maintenance, which is now typically achieved
through private firms. The APCC (2001) has argued that the effective and efficient management of
these assets is in the best interest of government, business and society (APCC 2001). As an
emerging field of endeavour, engineering asset management seeks to optimise the performance of
these engineering assets particularly the whole-of-life management of risks and expenditures for
the purpose of achieving organisational goals (British Standards 2003). Given the relative newness
of the field, much research is still needed in order to identify the optimal ways of procuring
engineering asset management and maintenance from the private sector by government (Ldre,
Austeng, Haugen and Kaklegg 2006).
Procuring engineering asset management and maintenance is a critical arena in which to conduct
research due to the size of expenditure involved in acquiring and maintaining these assets (APCC
2001), the typical longevity of the assets, and the significant risk posed to society if these assets
were to fail (Herder and Verwater-Lukso 2006). This paper argues that a richer understanding of
the procurement of engineering asset management and maintenance services can be achieved by
using perspectives from complex adaptive systems theory. The following sections outline the
administrative challenges faced by governments as they seek to arrange for the management and
maintenance of these assets, as these arrangements are of central interest to this research
project. Firstly, the complex public policy issues which have arisen due to new systems of
government in western democracies will be outlined, together with drawing out the implications of
these arrangements for engineering asset management. Theoretical perspectives which have been
deployed to explore this complexity will then be surveyed, and the utility of complex adaptive
systems (CAS) theory to investigate the procurement of engineering asset management will be
advanced.
Overview of public policy issues relevant to engineering asset management
The management of government services, such as engineering asset management, was
historically conducted by public agencies with in-house staff, occasionally supplemented by
external consultants, with ongoing maintenance undertaken for the most part by in-house
workforces (Hood 1991). In the early 1990s significant changes were introduced into western
democracies under the rubric of New Public Management (NPM) (Hood 1991). One of the main
changes which occurred under NPM was that those tasks which had been conducted previously inhouse were contracted-out to the private sector (Domberger 1994). Contracting-out, as it came to
be known, fundamentally changed the nature of service delivery for government agencies
(Verspaandonk, 2001), as government moved from delivering services directly, to contracting with
the private sector for the delivery of these services.
The introduction of NPM greatly increased the complexity of public services, as it required the
development of new sets of relationships between government and the private sector for the

delivery of specific services (Rhodes and MacKechnie 2003). Under NPM Government ceased to
provide many services directly to the public and instead engaged the services of private firms to
deliver, manage and maintain these assets and services on behalf of government (Rhodes 1997).
Osborne and Gaebler (1992) note that these new arrangements necessitated a shift in thinking
within government away from rowing (or doing the work) to steering (which involved managing
and guiding others who work on behalf of government). As a consequence of these changes,
governments of all persuasions are still attempting to find the optimal set of arrangements for
planning, tendering, contracting and managing the services of numerous private firms (Rhodes
1997), including those who have been contracted to provide the delivery and maintenance of
engineering assets (Ldre, et al. 2006). The process of sourcing goods and services by third
parties is typically referred to by government as procurement. As outlined by the APCC (2003, 2):
Procurement takes many forms and encompasses the acquisition of consumables
(goods); real property; capital equipment such as computers; built assets such as
hospitals, schools, roads and major facilities; and services such as office
accommodation, cleaning and security.
The interest of this paper is the procurement of engineering assets, which due to their size,
complexity, longevity and potential impact on society if mismanaged, involves a different set of
arrangements when compared to procuring smaller disposable items such as office supplies.
Procurement of engineering assets is in essence a series of decisions about the delivery system,
contract model and compensation format for the management and maintenance of a given asset
(Ldre et al. 2006). Numerous delivery systems exist in public policy documents which delineate
the overarching relationship between the contractor and government, such as managing
contractor, alliance contract, or period contract (New South Wales Government 2005). A variety of
contract systems exist which establish the range of services being procured from the private sector
for a particular asset, such as construct only; design and construct; or design, construct and
maintain (New South Wales Government 2005). Numerous funding options also exist. Having
determined the need for the asset, a department typically then justifies the need of the asset in
order to meet service obligations to society, and includes such plans in forward budget estimates
(e.g. Queensland Treasury 2003). The funding for the asset thus becomes part of intragovernmental negotiations over budgetary spending. The actual funding source for a specific asset
can be derived from a range of government and private sources (New South Wales Government
2005), as well as from a number of levels of government (Furneaux and Brown 2007b). Thus there
are several pathways available for the procurement of engineering assets including multiple
forms of contractual relationships and compensation formats possible in the delivery of
infrastructure procurement, with no single best method apparent (Ldre, et al. 2006). Such issues
are important as they relate to the efficiency and effectiveness of contractual and financial
mechanisms in achieving outcomes for government. Limited literature exists on the appropriate
types of procurement arrangements for particular classes or types of asset, and mechanisms
which are appropriate for specific situations, which this research seeks to address. As will be
outlined below, CAS theory has been advocated as providing a useful framework for the
examination of these complex sets of arrangements (Bovaird 2006).
The importance of ensuring engineering assets achieve their intended outcomes in an effective
and efficient manner, has meant that asset management practitioner literature has tended to focus
on the economic and technical decision making processes which are required for the effective
whole-of-life management of these assets (e.g. INGENIUM 2002), although attention to the whole
of life aspects of these large assets especially the disposal of these assets is only recently
being addressed (e.g. APCC 2007) . Engineering assets deliver critical services to the wider
community, such as the generation of electricity, provision of water, and national defence (Herder
and Verwater-Lukszo 2006), and the failure, neglect or sabotage of these engineering assets could
result in significant adverse outcomes for society as a whole (Hellstrm 2007; Godau 1999).
McIntyre and Pradhan (2003) and Goadau (1999) both argue that engineering systems such as
energy generation have for too long been considered as purely technological systems, and argue

that engineering assets include social, political, economic, and environmental factors. Herder and
Verwater-Lukszo (2006), argue for example, that engineering assets are not just pieces of
technology, but are really socio-technical systems consisting of both a social and a technical
element and note that the organisational and social context of engineering asset operations is
seldom investigated in research to date. This research project explicitly addresses this gap in the
literature by focusing on the range of organisations involved in making decisions concerning
engineering assets as these form key stakeholders who should be consulted when making
decisions concerning critical pieces of infrastructure (Mitchell, Agle and Wood 1997). As Bridgman
and Davis (2004) suggest, public policy arenas can be a chaotic place in which ideas must find a
path between the intentions of politicians, the interests of various government institutions, the
interpretation of bureaucrats and the intervention of pressure groups, media and citizens.
Research into engineering assets therefore needs to look beyond the economic and technical
aspects of procurement to the wider social, political and environmental context in which the asset
is situated. As CAS theory explicitly looks both at the behaviour of agents within a system, and how
the external environment of the system affects agents within the system, it would appear to be a
suitable framework for addressing the gap identified here.
In summary, engineering asset procurement is concerned with the planning, delivery and
maintenance of large assets which involves significant expenditure, and is integral to the
functioning of modern industrial societies (Herder and Verwater-Lukso 2006). Current practices
have emerged out of a raft of major changes in the public sector which meant that services once
provided by government on behalf of society were contracted out to the private sector to deliver on
behalf of government (Hood 1991). Procurement is in essence a series of decisions about an
engineering asset which should consider technical, economic, political and social factors (McIntyre
and Pradhan 2003). Many governments are still searching for the most appropriate way to procure
engineering assets and provide an optimal system of management and maintenance (Ldre, et al.
2006), as well as how to govern the ongoing relationships between government and private
industry (Rhodes 1997, 48) which are required for the effective delivery and maintenance of these
assets. As Jensen and Stonecash (2004, 22) note Despite its importance as a public policy issue
and the amount of research devoted to it, the determinants of successful public sector outsourcing
are still largely unknown. This paper will address this gap in the literature by examining the
procurement of engineering asset management and maintenance by government from private
firms.
Potential theoretical frameworks for exploring engineering asset procurement
Approaches prevalent in the academic literature relevant to the examination of procurement issues
in the public sector are principal-agent theory (Quiggin 1986) and network governance (Keast,
Brown and Mandell 2007). More recently, however CAS theory has been mooted as theory that
can extend network perspectives of government services (Klijn and Teisman 2007). These are
discussed further below.
Principal-agent theory has been a dominant theory in the analysis of government engagement with
construction and infrastructure industries. Much of the principal-agent theory literature has
focussed on the relationship between the owners and managers of firms particularly the
contractual relationship which is used to reduce risk, and to control opportunistic behaviour,
although it has also been applied to relationships between organisations (Eisenhardt 1989). As
government is a purchaser of buildings, construction and design firms are seen as the agents of
government, who is typically perceived as the principal (Quiggin 1996). Agency theory postulates
that people are self-interested at the personal level, and therefore have conflicts of interest in some
cooperative endeavours unless these relationships are mediated by arms length third party
transactions (Jensen 1994). Hence this theory has tended to focus on the specification of details in
contracts in order to prevent opportunistic behaviour by firms contracted to provide services to
government.
According to Eisenhardt (1989) principal-agent theory only represents a partial and adversarial
view of the world, and ignores a great deal of complexity in society, and in the business
environment. Part of the complexity is that government can have multiple roles in construction

projects such as assessor of infrastructure needs, project manager, facilitator, performance

sector, network planner, concession granter, inspector, contract manager, protector of the
environment, and representative of the public interest (Demirag 2004: 23). Increasingly funds for
major projects are drawn from multiple spheres of government, which means that contractors have
to meet the rules and regulations of two tiers of governments in order to be eligible to tender for
contracts (Brown, Furneaux, Janssen and Allan 2008). Further complexity emerges as multiple
departments may become involved in the planning and delivery stages of a project either directly
(Furneaux, Brown, McConville, McFallan, London and Burgess 2006a), or indirectly as multiple
outcomes are achieved from the procurement process, such as training, regional development and
even art delivered as a percentage of the construction project (Austen, Seymour, Brown, Furneaux
and McCabe 2007) or occupational health and safety (Brown et al. 2008). Simple notions of
principal and agent do not allow for the sheer number and diversity of roles that government can
play in individual engineering asset management projects. Eisner, Worsham, and Ringquist (1996)
argue that agency theory cannot be applied to complex modern bureaucracies as there are
simultaneously multiple agents and multiple principals engaged in the policy making process. In
their review of engineering asset management and maintenance Herder and Verwater-Lukszo
(2006, 119) argue that new and more intelligent methods are needed that are able to handle
multi-actor, multi-level, multi-objective and dynamic complexity of infrastructural operation.
Principal-agent theory has been held to provide significant explanatory power for exploring
government contracting, particularly in circumstances which involve fairly straight-forward, and
easily measured arrangements between government and contracted firms (Jensen and Stonecash
2004). Given the difficulty for principal-agent theory in dealing with a multiplicity of parties involved
in modern procurement arrangements (Eisner, Worsham and Ringquist 1996) such as that which
exists with engineering asset management (Herder and Verwater-Lukso 2006), alternative theories
are needed which can advance ways of understanding the interactions between the actors. One
such theory which has gained considerable attention in recent years is that of network
governance.
Considerable work has been undertaken in recent years to explore the new area of network
governance (e.g. Keast, Brown and Mandell 2007). Network governance is interested in the
networks of agents which are held to exist across most areas of government activity, including the
procurement of the management and maintenance of engineering assets (Osborn and Hegedoorn
1997). While NPM introduced delivery by the market of services for government, this did not mean
that hierarchy disappeared, as firms were still accountable to government and government was
accountable to society. Instead, in addition to hierarchy and market forms of organising the
economy, NPM resulted in networks of relationships between suppliers and government
developing (Keast, Mandell, and Brown 2006). Consequently hierarchy, market and network forms
of economic relationships co-existed in western economies (Rhodes 1997, 48), which resulted in
increased complexity due to overlapping of multiple policy domains (Keast, Mandell, and Brown
2006).
While the notion of networks, markets and hierarchies as modes of governance is not new, Rhodes
(1997) argues that it is the mix of these arrangements in public service contexts which matters.
Recent work has focused on the optimal mix of modes of governance in order to deliver services
and products (Provan and Kenis 2007), and the importance of different types of networks
(cooperation, coordination, collaboration) for different situations (Keast, Brown and Mandell 2007).
Network governance research has begun to advance understandings of how to influence network
structure, interaction and outcomes (Klijn and Koppenjan 2006; Provan and Kenis 2007), and the
extent to which these networks can be governed (Kooiman 1999). While principal-agent theory is
held to not cope with the multiple agents prevalent in the procurement of services by government,
Keast, Mandell, and Brown (2006) argue that network governance is able to accommodate multiple
agents and objectives prevalent in modern service delivery arrangements. Despite its power and
capability of demonstrating interactions between various agents in networked arrangements, Smith
and Stacey (1997) argue, however, that such approaches can overlook an important factor the
mixed governance modes of network, hierarchy and market in specific situations will not remain
stable but will adapt and evolve and tensions between mixed governance modes will generate new

forms of organisation. Klijn and Teisman (2007) agree, also noting that both the governance
system and its environment are constantly changing, and that complexity theory helps to explain
the way that agents, systems and environments interact and produce change. It is this area of
evolution, emergence and adaptation which CAS theory is held to provide a powerful alternative
perspective to network governance (White 2001). Pierce (2000), Blackman (2001), and Chapman
(2002) have likewise called for CAS research into public policy issues, arguing that this perspective
provides a highly useful alternative perspective to other theoretical approaches.
Network governance literature explores alternative modes of governing the sets of relationships
involved in the delivery of services to the network, thereby influencing the outcome of the network
(Klijn and Koppenjan 2006). Proponents of CAS agree with network theorists that hierarchical
approaches to management of complex delivery systems are generally not successful (Rhodes
and MacKechnie 2003). Part of the reason for this, according to CAS theory, is that order emerges
from the system itself, and that order cannot be imposed externally, as the system will always find
ways of adapting to such external controls and minimising their influence (Holland 1998). Thus
while a system and a network may appear similar, the means for creating order is perceived
differently.
Other authors researching in the area of engineering asset management argue that it is imperative
to examine engineering assets from a CAS perspective due to the inter-related nature of the
systems, and their importance to modern societies (Rinaldi, Peerenboom, and Kelly 2001). Price
and Akhlaghi (1999) and Herder and Verwater-Lukszo (2006) have also argued that engineering
asset management needs to be examined from a CAS perspective in order to address the
demands of multiple stakeholders. This is achieved by examining the range of agents involved in
the system. Godau (1999) argues that traditional approaches to managing engineering assets do
not address the conflicting needs of technical, economic, managerial, environmental, political and
social agents, and argues that a CAS perspective is needed to take into consideration all of these
agendas. An improved understanding of the dynamics of these systems is likely to lead to a better
understanding of how to manage the problems encountered within these systems (Smith and
Stacey 1997).
Despite all of these calls, Richardson (2006) suggests that CAS as a theory has only newly been
applied to real world public policy problems. Klijn and Teisman (2007) agree arguing that there
have been relatively few applications of CAS theory to public policy problems. These perspectives
are summarised by Parsons:
The application of complex adaptive systems to social systems is fairly recent
and appears to hold promise for helping us to work within the complexity of
todays world (Parsons 2007, 407).
Published reports using CAS as a theoretical approach have been found in a variety of public
policy areas including: community involvement (Midgley and Richardson 2007), drug trade (Coyle
and Alexander 1997), federalism (Coghill 2004), health care and public health (Bar-Yam 2006;
Vennix and Gubbels 1992; White 2000), international development agencies (Smith and Stacey
1997), occupational health and safety (Koppenjan 2001), security services (Meek de Ladurantey
and Newell 2007), sustainable development (Daneke 2001; Moxnes 2000) and tobacco excise
(Cavana and Clifford 2006). Each of these studies have begun to deploy CAS theory as a
framework for examining specific public policy problems. CAS theories have also been applied to
marketing (Markose 2005; Wollin and Perry 2004) and economics (Robson 2005). Thus while CAS
theory is relatively new to public policy research (Klijn and Teisman 2007) numerous authors are
now beginning to apply this theoretical framework to explore a variety of public policy issues. Such
calls are not restricted to investigation of general policy matters, but have also emerged specifically
in the area of government procurement.
While CAS theory has been held to provide a very useful framework for evaluating the
appropriateness of policy initiatives in general (Sanderson 2000), some researchers have also
used CAS theory to explore public procurement. One study in the United Kingdom focuses on the

various forms of market relationships possible in procurement of government services (Bovaird

2006). Bovaird (2006) argues that CAS theory is a useful theoretical perspective to understand
procurement as it provides a framework for examining multiple agents, their relationships, the rules
they utilised to make decisions and how these decisions can affect the system overall. Hitchins
(2003) also argues that CAS theory is useful in understanding how government procurement
arrangements can affect the market, and in turn are affected by the market. Klijn and Teisman
(2007) in their exploration of CAS theory use a case study of the procurement of rail management
and maintenance services in the United Kingdom. They found the theory useful in identifying the
key agents in the system (government and private) and the impact of decisions made by these key
agents on the capability of the system (rail) being managed.
In summary, engineering asset management is a significant area of research, as engineering
assets involve considerable expenditure by government, and are of critical importance for the
effective functioning of modern societies. Such arrangements involve complex sets of decisions
concerning funding, contracting, and relationships, (Ldre, et al. 2006) and involve the
management of multi-agents and multi-objectives (Herder and Verwater-Lukszo (2006, 119).
Numerous authors (Rinaldi, Peerenboom, and Kelly 2001; Price and Akhlaghi 1999; Herder and
Verwater-Lukszo 2006; Godau 1999) have argued that engineering assets should be investigated
from a complex adaptive perspective. Bovaird (2006) has specifically argued that CAS theory
provides a very useful perspective for the examination of government procurement.
Engineering asset procurement is an important area which appears to have not been studied
extensively from a CAS perspective to date. This research seeks to expand current literature on
the procurement of engineering asset management, by using a CAS perspective, which numerous
authors argue is important (Smith and Stacey 1997) and will result in improved understanding and
therefore improved management of engineering assets.

Area and topic

This paper will develop and validate a framework for analysing engineering asset procurement
from a CAS perspective, which takes into consideration both conceptually and methodologically
the complexity associated with the procurement of engineering assets. In short it seeks to develop
a richer understanding of engineering asset procurement using CAS theory as a theoretical
framework. A better understanding of the dynamics of such a system should in turn result in better
management, maintenance and long-term decision making for asset management.
The topic is thus situated in the broad area of public policy. However, a CAS perspective will be
utilised to examine a public policy issue: engineering asset procurement.

Literature Review
This section provides an overview of CAS theory, and how this theory provides a useful conceptual
framework for examining public procurement of engineering asset management services. Having
initially explored CAS, together with some challenges involved with the deployment and
operationalisation of the theory in management contexts, specific research questions, problem
specification and significance of the study are detailed.

Complexity and Complex Adaptive Systems (CAS)

Complexity is concerned with phenomena which cannot be easily explained by positivistic scientific
methods (Haynes 2007). As has been noted above, CAS is a promising line of research for this
study as it provides a framework which accommodates the complexity of public sector procurement
(Bovaird 2006), the engineering assets which are procured (Herder and Verwater-Lukszo 2006),
and the emergent, changing nature of modern governance arrangements (Klijn and Teisman 2007;
Meek, de Ladurantey and Newell 2007) which are required to care for such assets.
Systems thinking is a way of understanding and investigating complexity, and CAS theory is now
regarded as a strong candidate for providing a unifying notion of complexity (Chu, Strand and

Jelland 2003, 19). This paper will use CAS theory as a way of initiating and guiding thinking about
the procurement of engineering asset management (Richardson 2005a). This is in line with other
public policy research which posits CAS theory as providing a framework for analysis that is
beyond a mere metaphor, and instead that it provides a conceptual framework, a way of thinking
and a way of seeing the world (Mitleton-Kelly 2003, 26).
There is no one theory concerning CAS (Mitleton-Kelly 2003), but rather several theories arising
from the studies in a variety of sciences. The following definition will be used in this paper:
A complex adaptive system consists of a large and diverse
number of agents that interact in nonlinear and adaptive
ways. In a densely intertwined web of interacting agents
each agent is responding to other agents and the
environment as a whole; it is continually adapting in the
context of its relationships with other agents (Parsons 2007,
406).
While there are various perspectives on complexity, this paper will utilise CAS as a conceptual
framework for examining engineering asset procurement.
While application of CAS theory to public policy arenas has been held to improve the
understanding and performance of organisational systems, considerable work is needed to apply
this concept to organisational contexts in general (Anderson 1999), and to public service
arrangements, in particular (Daneke 2005).
Difficulties in applying CAS to procurement of engineering assets
Whetten (1989) argues that bringing a perspective into management from another field can be a
fruitful way of developing new theoretical understanding. It is important to note in this regard that
CAS theory originated in the fields of physics and biology. In physics for example after three
hundred years of dissecting everything into molecules and atoms nuclei and quarks, [scientists]
were starting to look at how those pieces go together into complex wholes (Waldrop 1992, 16).
Likewise in biology where people had spent the last twenty years laying bare the molecular
mechanisms of DNA, and proteins, and all the other components of the cell. Now they were also
beginning to grapple with the essential mystery: how can several quadrillion such molecules
organize themselves into an entity that moves, that responds, that reproduces, that is alive?
(Waldrop 1992, 16). Thus, CAS theory, which has its origins in physics, chemistry, biology and
computational studies, may well provide novel and interesting perspectives which can improve our
theoretical understanding (Whetten 1989), provide significant advancement in our understanding of
procurement (Bovaird 2006) and provide insights into engineering assets such as rail (Klijn and
Teisman 2007).
Mitleton-Kelly (2003) sounds a note of caution however, arguing that while the elements of
systems of atomic particles may bear some resemblance to activities in organisations, there are
also critical differences which need to be acknowledged particularly the ability of people in
organisations to reflect, learn and behave in ways that are not entirely predictable. Griffin and
Shaw (2000) agree, contending that systems thinking is inappropriate for modelling the activities of
human systems, as humans can choose between a variety of options, and their actions cannot be
predicted with absolute certainty. Ball (2005) is more optimistic arguing that CAS thinking can
provide a framework which enables comprehension of non-linear relationships typical in complex
systems.
Introna (2003) has suggested that a way forward to the application of CAS theories to
organisational research is to begin with some of the insights from the parent domain, and examine
their applicability through empirical work, which could lead to novel understandings of social
systems, and possibly ways of intervening in them. Eisenhardt (2002) argues that such constructs
would still need to be tested in order to demonstrate their validity. Daneke (2005) agrees, noting
there are significant opportunities for the application of systems thinking to public administration.

As noted above, CAS theory has been advocated as an important perspective which can be used
to develop a better understanding of engineering assets (Smith and Stacey 1997), and government
procurement (Bovaird 2007), and has been demonstrated as holding utility for the interaction
between government procurement of engineering assets and the market (Hitchins 2003). These
need to be tested, particularly given the relatively recent application of the theory to public policy
problems (Klijn and Teisman 2007). As will be discussed later, the utility of CAS theory to
understand government procurement of engineering asset management will be explored through a
number of case studies.

Problem
There is considerable complexity involved in the procurement of engineering assets, and with the
increased attention being paid to these assets by the public, government and industry alike, a
study of the elements of the procurement system is timely. In addition, while various studies have
examined specific aspects or elements of the procurement system, relatively few have examined
procurement in its entirety from a CAS perspective. This study will be of value to governments
seeking a framework for making decisions concerning engineering assets, for industry seeking to
engage with government, and for professional associations and education institutions which may
wish to train staff in decision making processes under conditions of deep uncertainty.
Numerous authors have argued that CAS theory provides a powerful analytical device for
examining and exploring complex phenomena (Parsons 2007, 407) for public policy issues in
general (Klijn and Teisman 2007; Meek de Ladurantey and Newell 2007; Richardson 2006)
engineering asset management (Rinaldi, Peerenboom, and Kelly 2001; Smith and Stacey 1997;
Price and Akhlaghi 1999; Herder and Verwater-Lukszo 2006; Godau 1999), and the procurement
of services such as engineering asset management (Klein and Teisman 2007). The application of
CAS to social systems however, is fairly recent and there are many disagreements about how to
apply these concepts (Parsons 2007). Considerable work is needed to apply CAS theory to
organisational systems, particularly the operationalising and modelling of such concepts (Anderson
1999), and the testing of the theory in multiple cases in order to demonstrate and develop the
application of the theory to public policy (Daneke 2005).

Statement of purpose
The purpose of this study then is to examine the procurement of engineering assets from a CAS
perspective. Such analysis will seek to identify the agents, their roles and interaction; the rules
which underlie the system, and the mechanisms by which such systems change and adapt, and
are affected by their environment. Such research is exploratory (Babbie 2004) and is the first stage
in developing a formal model of the procurement system of engineering assets. The purpose of this
activity is in order to test the validity of CAS theory as a suitable framework for understanding the
behaviour of government procurement from a systems perspective. By demonstrating the utility of
CAS to explore the procurement of engineering asset management, the management,
maintenance and long-term decision making for asset management is enhanced.
Given the relative newness of this perspective to organisational studies in general, and public
policy in particular, the research will demonstrate the utility of systems thinking by examining a
specific area of government activity (procurement) in a specific area of this activity (engineering
assets).

Operationalising CAS research in public policy contexts

As noted earlier, an important process in developing an understating of a CAS is to develop a
model of the system (Holland 1998). The development of a model of a public policy problem as a
CAS, has been acknowledged as posing considerable difficulty (Rhodes and MacKechnie 2003).
Chu et al. (2003, 27) agree arguing that the sheer complexity of a public policy problems, and the
number of agents involved, makes the modelling of public policy areas as CAS very difficult to
model.

As a way forward around this dilemma, and in order to apply the concept of CAS to procurement of
engineering assets, it is prudent to firstly detail the elements, components and principles which are
held to be part of all systems. This approach is particularly important if the CAS is used as a
framework for thinking and analysis, not merely as a metaphor (Mitleton-Kelly 2003). As a first
step, the core elements of a CAS must be described, and how this might be operationalised in this
research project.
Firstly what is a public policy system? Rhodes and MacKechnie propose that:
Public service systems consist of multiple organizations engaged in the provision of a
specific set of goods and services that are of value to the majority of consumer-citizens
(Rhodes and MacKechnie 2003, 61).
In order to apply the concept of CAS to procurement of engineering assets, it is prudent to firstly
detail the elements, components and principles which are held to be part of all systems. As a first
step then, the core elements of a CAS must firstly be educed. Once this has been achieved, an
initial outline detailing how such elements might be operationalised in examining the procurement
of engineering assets is undertaken.
Given the newness of the field (Parsons 2007) it is important to build on and expand existing
empirical and theoretical foundations. Consequently, this paper identifies and those elements
which are common in most papers on CAS as applied to organisations and public policy systems
(eg Anderson 1999, Mittleton-Kelly 2003, Rhodes and Mackechnie 2003, Anderson et al, 2005,
Klijn 2008). These are:
o
o
o
o

Agents who interact according to schemata

Self-organisation (also termed emergence)
Co-evolution
Adaptation, evolution and recombination

These elements are discussed in more detail below:

Agents who interact according to rules (schema)
Rather than approaching complex systems by reducing them to a set of causal variables, CAS
models can show how complex outcomes flow from the interaction of agents based on a set of
simple rules or schemata (Anderson 1999). The identification of agents is considered the first step
in any CAS study 1 (Holland 1995). For organisational researchers, agents can be identified as
individuals, groups, or coalitions of groups (Anderson 1999, 219).
For this research it is proposed that the main agents in a procurement system can be largely
segmented into two groups according to their influence on the decision making process direct
and indirect. Direct agents include government agencies involved in the procurement process, and
the firms involved in the planning, delivery and maintenance of specific assets. Indirect agents
would be those who seek to influence the decision making process, but have no direct involvement
in the decision itself. These include policy advisors, and political lobby groups. In terms familiar
with public policy literature, these would be known as policy networks and delivery networks. An
incomplete list is provided below:
Table 1 - Initial list of actors involved in procurement systems

Government
Procurement units
Clients
Politicians

Direct
Private
Principal contractors
Sub contractors
(many of these are on pre-

Government
Policy officers
Ministerial advisors

Indirect
Private
Policy networks
Advocacy groups
Lobby groups

1 Some authors in the literature use the term actors to refer to individual elements of a Complex Adaptive System. This paper will use
the alternative term agent instead, as often individuals act as agents for their organisation.

qualified supplier databases)

A complete list would require full analysis of prequalification schemes, and research methods for
eliciting the key agents in the system, such as snowball sampling in interview. The interaction
between these agents is determined according to the rules of the system which are discussed in
the next section.
Interaction according to rules
Often an approach to studying organisations is to identify independent and dependant variables at
the same level of analysis in order to demonstrate cause-and-effect relationships. CAS theory
approaches this differently, by asking how changes in the rules of agents, or the interaction
between agents, result in outcomes for the system as a whole (Anderson 1999). That individuals
use rules to make decisions is reflected in the notion that agents have frames of reference or
schemata (Rhodes and MacKechnie 2003) by which they interpret and evaluate information (Kiljn
and Teisman 2007). These schemata include an understanding of how the world, people,
organisations and procedures work (Wolf 2005, 187).
Rules provide codes of meaning that facilitate the interpretation of ambiguous
worlds. They embody collective and individual roles, identities, rights and
obligations, interests, values worldviews, and memory, and thus constrain the
allocation of attention, standards of evaluation, priorities, perceptions and
resources (Olsen 2005, p.9).
From a symbolic interaction perspective individual agents struggle to control shared interpretation;
roles and rules are negotiations and gambits in the struggle to define and construct meaning
(Anderson 1999). For leading CAS authors such as Gell-Mann (1994) it is precisely this capability
of being able to create schemata that distinguishes CAS from other complex systems such as
galaxies. In fact, CAS can have many competing schemata which compete against each other
the ones that prove to be most salient are the ones that are reinforced (Holland 1995). From a
public policy perspective:
making sense of contemporary public administration then, requires and understanding of
the complex ecology of institutions, actors, rules, values, principles, goals, interests, beliefs,
powers and cleavages in which it operates (Olsen 2005, p7).
Holland (1995) contends that rules can be classified into two main types rules which regulate the
action of agents, and rules about the system itself. This point is echoed by Klijn (2001, 2007)
suggesting that in public policy systems there are rules which focus on the policy arena itself
(arena rules), and those which relate to the interaction of agents in a network (interaction rules).
This can be demonstrated in the following table:
Table 2 Main types of rules in policy systems (adapted from Klijn 2001, 2007)

Interaction rules

System rules

Description
Rules which regulate
agents interactions

Rules which regulate

the setting or policy
arena

Aspects
Access to policy game
(who may enter the
game, exit options,
etc)
Interaction in policy
game (what is
permitted or not in
interactions)
Reality (what agents
consider as core
business or quality)
Payoff (costs and
benefits for agents)

Examples
Pre-qualified supplier
arrangements
Institutional
arrangements for
procurement
Conflict
Information

Identity of agents
Product rules
Status
Evaluation criteria

Positions (positions of
agents and relations
between positions)

Status
Power

Like other elements of CAS theory, the notion of agent decision making based upon rules is not
totally new to public management, as March and Simon have argued (1993, 8):
The matching of rules to situations rests on the logic of appropriateness. Actions
are chosen by recognizing a situation as being of a familiar, frequently
encountered, type, and matching the recognized situation to a set of rules The
logic of appropriateness is linking to conceptions of experience, roles, intuition and
expert knowledge. It deals with calculation mainly as a means of retrieving
experience preserved in the organizations files or individuals memories.
Determination of the rules which guide the activity of agents would typically require interviews with
the agents of the system. Applying this to procurement then requires an investigation of the rules
about the procurement system and the rules of interaction within the system. As noted in the
introduction there are a large number of decisions which need to be made in any procurement
activity. Two key sets of rules relate to decisions concerning the institutional arrangements
(interaction rules) involved in delivering the engineering asset, and in decisions concerning the
engineering asset itself (system rules).

Interaction rules in procurement of engineering assets

As noted above, the determination about which agents are permitted to participate in a given arena
is governed by a range of rules about the system itself. Information concerning the role
government agencies are permitted in the procurement process would be elicited from policy
documents and interviews. The determination of which private agents are permitted to tender for
engineering procurement would be elicited by examining policy documents, interviews and various
mechanisms such as prequalification schemes.
System Rules in procurement of engineering assets
In terms of decisions concerning the engineering asset itself, there are three well recognised and
competing rules for engineering assets: price, quality and cost. These rules are seen as
competing, and project management literature argues that all three requirements cannot be
achieved at any given time. For example, if the decision is made to deliver a gold-plated solution
(one of very high quality) then time and cost are likely to increase. Likewise, if time is of the
essence, then quality and cost are likely to be affected. As noted above, validation and elicitation
of additional rules would require interviews with key informants.
According to Holland (1995), from a relatively simple set of rules, a surprising amount of diversity is
possible within a system. So establishing the key rules which influence procurement decision
making are an important part of this research project. However, rules for the procurement of
engineering assets (such as price, quality and time) can be seen to be in tension with each other.
Agents must make choices between these alternatives, which is discussed in the next section.

Choosing between alternatives (also termed the fitness function)

As noted above, CAS theory argues that there are often competing rules in a system, and agents
must choose between them. Choosing between alternatives is determined by what is termed the
fitness function. Fitness functions govern how the agent will choose among alternative actions
(Gell-Man 1984; Holland 1998). The most appropriate rules, (i.e. the ones that work) are the ones
that tend to be reinforced as they have a high level of fitness for agents in relation to the
landscape. A biological example is the choice of fight or flight for animals the most appropriate
choice is determined by what threat the organism is facing. This concept of a fitness function is
similar to the logic of appropriateness outlined by March and Simon (1993), as organisations and
individuals make choices based on past experience and learning about what the most appropriate
action would be in a given circumstance. This concept of learning and reinforcement of rules is
explored in the section on adaptation below.
In procurement, little research has been undertaken to date to understand how individual agents
choose between differing procurement alternatives in government, with preliminary work
suggesting that decision trees are employed to make various procurement choices (Ldre et al.
2006). In this project, having determined the agents in the procurement system, and the rules they
operate by, how the agents make choices between alternative procurement methods needs to be
elicited, which is most likely possible through semi-structured interviewing. Thus the first key
element of examining procurement of engineering assets from a CAS perspective is to identify the
agents, their rules and how they choose between these rules.
CAS theory holds that the order in a system emerges from the interaction of agents based upon
rules which are outlined in the next section.
Self-organisation (also termed emergence)
The structure and dynamics of a CAS are a result of choices by the agents, as they learn and
adapt to actions of other agents (Albino et al., 2005). In other words there is no formal order
imposed from outside of the CAS, and order emerges from the interactions between the agents at
a local level (Stacey and Griffin 2005, 7). Emergence is the term used in CAS theory to describe
the phenomena of patterns at a higher level of abstraction that arise from interactions among lower
level agents (Rhodes and MacKechnie 2003, 63).
That a complex process can be self organising is not new in and of itself. Adam Smith (1971)
introduced the notion of the invisible hand of capitalism where markets were efficient yet were
governed by laws of supply and demand, not government. What is relatively new is the application
of this concept to management (Clippinger 1999, 2). Klijn and Teisman (2007) argue that notions of
self-organisation have parallels in institutional theory with its idea that structures emerge from
interaction of agents, as they interpret and use institutional rules. CAS models are inherently multilevel as the order is seen as an emergent property which results from lower levels of aggregate
behaviour (Anderson 1999). No application of CAS to public policy has yet been able to
demonstrate this multi-level nature of CAS to date. This research project will explicitly attempt to
identify the multiple levels involved in engineering assets procurement as part of each case study.
Daneke (2005, 95) argues that the primary feature of social systems thinking is its focus on those
elements that emerge from the interactions of agents and institutions. Emergence is seen to be
both a product of structures and relationships. For example, Auyang (1998, 176) argues that
emergent characters mostly belong to the structural aspects of the systems. Holland (1998, 121122), on the other hand, argues that emergence is above all a product of coupled, context
dependant interaction. Technically these interactions, and the resulting system, are nonlinear. The
behaviour of the system cannot be obtained by summing up the behaviours of its constituent parts
unless we take the nonlinear interactions into account. Consequently, emergence must study
the analysis of the parts as well as the interactions between the parts as we will not understand
these complex systems until we understand the emergence phenomena that attend them (Holland
1998, 4).

The structure of agents interacting in governmental arenas has been of growing concern to public
policy researchers particularly those research the various modes of governance: hierarchy,
network and market (Brown and Keast 2007). Boisot and Child (1999) argue that these different
organising arrangements are the main mechanisms 2 by which agents in CAS cope with
complexity. After many years of research and theorising, markets, hierarchies and networks have
been accepted as the fundamental, different modes of organising (Rhodes and MacKechnie 2003).
Rhodes (1997) argued that the effective delivery of government services relies on ensure the right
mix of these arrangements. Unfortunately, the right mix for differing circumstances is difficult to
identify. In this context, network arrangements are seen to be informal shadows of formal systems
(Smith and Stacy 1997). The interaction between the formal (hierarchy) and informal (network)
systems produces emergent order, which may or may not be in line with the intentions of those in
authority (Smith and Stacey 1997). As Meek et al. (2007, p.24) have argued administrative
networks, shared governance, and co-production of public services developed in the conjunctive
state, are real-world examples of the emergent properties of complex adaptive systems.
The implications of the concept of emergence for this research project are that lower levels of
interaction result in order at the same level and also higher levels of order. Consequently the
interaction of agents is held in CAS theory to result in higher levels of order and this emergence of
order should be explored as part of the research project.
Emergence of higher order structures in the procurement arena have emerged in Australia, with
the formation of groups such as the Australian Procurement and Construction Council (APCC) and
the Australian Asset Management Collaborative Group (AAMCoG). While the APCC does not have
executive powers, it provides a forum for knowledge sharing, intelligence gathering and has the
information networks to draw on for formulating solutions (APCC 2008), and was established
following interaction from state and territory jurisdictions. Similarly AAMCoG exists to facilitate
collaboration and knowledge sharing in the area of asset management (CIEAM 2006). Tilebein
(2006) advanced a cogent explanation of the phenomena of emergence in organisations while
studying supply chain management. An adaptation of Tilebeins (2006) model is given in Figure 2
below, and depicts emergent order as applied to government agencies and private firms and
associations involved in procurement.
Figure 1 Emergence of order in organisational contexts (adapted from Tilebein 2006, 1097).

Consequently the case studies would also need to explore the notion of emergence in the
interviews seeking to identify higher levels of order, often new institutions or organisations which
have emerged from interactions at lower levels of action.

Boisot and Child (1999) use the terms fief and clan in stead of the term networks. The term network has been retained as it is the
term commonly used in a range of literatures (e.g. Rhodes 1997; Keast, Brown and Mandel 2006).

Co-Evolution
In a CAS, agents are seen as being interconnected so that the behaviour of an agent is influenced
by the behaviour of other agents in the system. As one agent changes, so does the other hence
the understanding of co-evolution. It is this interconnectedness of agents which distinguishes CAS
models from other systems models. For example in systems dynamic models, variables are
connected to each other by feedback loops; in CAS models, agents are connected to one another
by feedback loops (Anderson 1999, p. 219 italics in the original). The notion of co-evolution has
particular purchase for decision making in networks where experiences and choice influence each
other because of learning processes (Klijn and Teisman 2007, 9). Morgan (1997) refers to this
process as mutual causality and provides and example in the mad cow problem which decimated
an industry in Britain.
Figure 2 How decisions by one agent influence other agents decisions (Morgan 1997, 279)

The idea that agents, their rules and their interactions co-evolve in CASs has been outlined above.
One way of applying this to procurement system is that in organisations and markets, the least-fit
element of systems tend to be eliminated organisations replace their least efficient members, and
least efficient firms in an industry tend to go out of business (Anderson 1999). A new agent drawn
at random is likely to have a higher average fitness that the weak one replaces, which sets of
changes in relationships between agents and can cause a cascade of changes in co-evolutionary
adaptation in the system. With new actors, or new rules, or new relationships between actors, this
causes a cascade of changes in the system.
Consequently, changes to the participants in a system, or their rules would need to be examined
through interviews and reference to secondary data such as prequalification schemes in order to
explore the concept of co-evolution.
Adaptation, Recombination, and Evolution
Adaptation in CAS occurs due to changes in the environment, the choices of agents and often a
dynamic feedback between these two. When the environment of the system changes, so does the
behaviour of it agents and as a result, the behaviour of the system as a whole in other words
the system learns and adapts to the new environment (Lewin and Regine 2003). CAS also evolves
over time through the entry, exit, and change of agents, as well as changes in the linkages

between agents (Anderson 1999). The structure and dynamics of a CAS are a result of choices by
the agents, as they learn and adapt to actions of other agents (Albino et al., 2005). Chapman
(2000) argues that learning is a critical issue which is enabled by systems approaches to policy
problems.
Research into engineering asset procurement from a CAS perspective therefore needs to examine
how the procurement system (the agents, their relationships, and/or the rules) have changed over
time, due to either the choice of agents, or changes in the environment.
In order to demonstrate the concept of self-organisation for procurement policy is perhaps best
illustrated by drawing on extant research of the author, together with the underpinning adaptation
and co-evolution of systems. In Furneaux et al. (2006) a summary of variant procurement
approaches for public works in Australia was advanced. In the course of completing this research,
it was evident that the procurement system in many states had recently changed from that which
followed the widespread introduction of contracting out in Australian jurisdictions, and this was
largely driven by interaction with construction firms and client agencies, or by changes in the
environment. Figure 1 provides and initial conceptualisation of this:
Figure 3 - Initial conceptualisation of Policy Emergence

Procurement Policy

Interaction between agents

Provides rules
Procurement of specific
assets

Alters rules
Environmental
factors (eg limited
labour, resources
or companies to
deliver)

Summary
Following Rhodes and MacKechnie (2003) it is proposed that an investigation into CAS in public
policy might be operationalised by firstly undertaking a description of the agents of the system.
Identifying who the agents are, and the nature and level of connections between agents, are key
building blocks to understanding the nature of a system and how it operates and helps to identify
the state of the system (Holland 1995). This is likely to be achieved by starting with the list of
suppliers who are pre-qualified to deliver a particular type of engineering asset, and the
government agencies which are responsible for the procurement of such assets. The next step is
to examine the schemata (or rules) by which the agents make decisions, the way agents decide
which decisions to take notice of, and the process by which rules change (Rhodes and
MacKecknie 2003). An example of this might be under what circumstances an alliance
methodology is chosen for construction of an asset. Investigation of policy documents suggests
that alliance contracts are often used in highly complex infrastructure projects such as dams (New
South Wales Treasury 2005), although this does not explain why private firms would decide to
participate in such arrangements. A related activity is to examine how agents choose between
different options (Holland 1995). An example of this might be to ask agents how they decide to
tender or participate in design-construct arrangements, as opposed to designconstruct-maintain
arrangements. The final steps would be to identify how the system acts in relation to other systems
in the environment, and how the system adapts and learns over time (Smith and Stacey 1997). An
example of this is the recent changes of many smaller government agencies who recently
centralised procurement in response to changes in the environment (Furneaux, Brown, Allan,
McConville, McFallan, London and Burgess 2006b).
These key areas are stated below as formal research objectives.

Research aim and objectives

o

The aim of this research is to use CAS theory as a framework to develop a richer
understanding of engineering asset procurement systems.

In order to deploy CAS theory in this context, it is necessary to identify the key elements of such a
system and how they operate. As outlined in the literature review above, these include:
o Identifying the central agents of the engineering asset procurement system
o How do these agents elements interact
o The rules which guide decisions made by these agents
o The tension(s) between rules and agents
o How are these tensions resolved? (which one wins and why)
o Changes (emergence) in the engineering asset procurement system to changes in
the environment or decisions of agents?

Significance
Yin (2003b) suggests that there can be three areas of significance for research: contribution to
knowledge in the area, to policy and to practitioners.
To knowledge in the area
Various authors have called for the application of CAS perspective to understanding policy
processes (Bovaird 2007; Klijn and Teisman 2007; Meek, de Ladurantey and Newell 2007;
Richardson 2006; Sanderson 2000; Stewart and Ayres year) although empirical investigations are
few.
Through Complex Adaptive System models and empirical verification, the
black box of public service systems may become more transparent, thereby
progressing our understanding of the link between agent decisions, agent
interactions and system outcomes (Rhodes and MacKechnie 2003, 80).
By demonstrating the utility of CAS theory in explaining the outcomes of procurement systems, an
alternative theoretical model of policy development processes within government can be
developed.
To policy
The implications for policy are far reaching. Many procurement approaches assume simple
relationships between principal and agent, and consequently may not take into consideration
adaptation and change in the system itself. A better understanding of the procurement of
engineering assets is likely to lead to a better understanding of how to manage the problems
encountered within these systems (Smith and Stacey 1997). While not seen as an intervention in
and of itself, by developing a model of the system it is possible this can lead to changes at a policy
level, as the influence of various forces becomes apparent (Wolstenholme 1992). Mass (1986) has
argued that models of system are sufficiently accurate, they can move decision makers towards a
deeper conceptualisation of the impact of their policy choices, and thereby lead to a change in their
choices. Improved procurement choices should result in enhanced economic and social benefits
for stakeholders.
To practice
Similar to the policy implications, an improved understanding of the nature and dynamics of a
system can lead to improved outcomes for participants as assumptions are tested and improved
and improved understanding can lead to better ways to address problems encountered in the
system (Smith and Stacey 1997). As noted in the problem section above, some of the key
problems affecting the procurement system at the moment are rising costs, and lack of skilled
labour. The private sector should also benefit as depiction of the procurement system will enhance
their participation as agents in that system particularly the rules by which the system operates,
and how change occurs in the system.

By examining engineering asset procurement from a CAS perspective, it is thus anticipated to

advance theoretical and practical knowledge of how such arrangements can be governed and
effectively managed.

Methodology
As noted in the preceding sections, this research project seeks to apply CAS thinking to research
into the procurement of engineering asset management. Given the note of caution raised by
Mitleton-Kelly (2003) concerning the importance of testing the applicability of the assumptions of
CAS theory to public administration the overarching methodological approach to the research will
be that of a series of qualitative case studies, which will examine separate examples of the
procurement of engineering assets. Each case study will answer the research objectives outlined
above by seeking to identify each of the agents in the system, how they interact with each other,
the rules which underpin their behaviour how the system adapts and changes, and if higher order
emergent structures can be identified. In order to explore these elements, initially a series of
interviews will be undertaken with key informants to identify their goals, rules, and interactions with
other agents. From these initial sets of interviews, initial models of the system will be developed. A
series of focus groups with key informants will then test and refine these models, in order to
explore the functioning of the system as a whole, particularly interaction, adaptation and feedback
within the system, and how the system and its rules change over time. Once each of the case
studies has been completed, cross case comparisons can occur to elicit similarities and differences
between the cases. These methodologies are discussed in detail below.

Case studies
Case studies provide for in-depth analysis of a particular issue or technology as it impacts an
organisation or industry, and can provide strong recommendations for improvements in theory,
technology or policy. Case studies in the area of policy have been called for as a way of advancing
public policy practice (Osborne and Brown 2005, Stake 2005). Given the dearth of existing work
undertaken in CAS research for public policy (Klijn and Teisman 2007) an exploratory case study
methodology (Babbie 2004, 8788) is appropriate. This is because a series of case studies can
develop the application or understanding of theory to an area (Eisenhardt 2002) such as has been
advocated for CAS and public policy (Daneke 2005).
A case study is a method for learning about a complex instance, based on a comprehensive
understanding of that instance obtained by extensive descriptions and analysis of that instance
taken as a whole and in its context (U.S. General Accounting Office 1990, cited in Mertens
2005:237). The objective of conducting multiple cases is to examine particular phenomena in
multiple contexts in order to understand how, when and why it occurs (Huberman and Miles 1994).
Multiple cases can enable powerful explanation of a particular process if the research is carefully
ordered (Huberman and Miles 1994). Additionally, the multiple case studies enhance the
generalisability of findings back to theory or policy (Yin 2003a), due to the fact that the same
phenomena are investigated in multiple contexts (Schofield 2002).
Therefore multiple case study approach for examining systems such as the procurement of
engineering assets, has both been called for by leading authors and appears appropriate. Such an
approach is held to have utility due to the lack of research in the area and have been explicitly
called for as a way of processing understanding of complex adaptive systems in general (Holland
1995) and also in the application of CAS theory to specific industry contexts (Herder and VerwaterLukszo 2006). A multiple case study framework will enable a better understanding of the
phenomena of engineering asset procurement from a CAS perspective.

Sample
Sampling within qualitative research is not necessarily based on random sampling, as often a
specific instance or example is being examined in detail. For this project, selection of cases will be
theory based as the issue is to find examples of policy situations in which CAS theory can be
used to elaborate and examine these cases (Miles and Huberman 1994).
In order to gain access to specific cases, the partners of the CRC funding this research project, the
Centre for Integrated Engineering Asset Management (CIEAM), will be approached initially. These
include:
o Defence: ASC, DSTO,
o Electricity: Delta Electrical, Electrical Supply Authority
o Water: Sun Water
o Other partners: Queensland Rail, Rio Tinto, ANSTO, Commerce Queensland.
Given that the subject of the cases are engineering assets, it is quite likely that different
methodologies or processes are in place for procuring different types or classes of assets. For
example, evidence suggests that the procurement of dams in South East Queensland is almost
always delivered via an alliance contract, whereas the procurement of public buildings would not.
Major roads contracts are often delivered via competitive tendering arrangements, although
alliance contracts have also been used in specific instances; whereas smaller maintenance
contracts are almost always via sole invitee arrangements (Queensland Department of Main
Roads 2006). For larger assets such as certain military equipment or types of power station, it is
also likely that there are very few suppliers able to tender for specific types of assets, which may in
turn affect the procurement process. Consequently, cases shall be selected in order to represent a
number of different asset classes (such as defence, transport - roads and rail, and utilities - water
and electricity). As examples of case studies would be the procurement and establishment
processes involved in the South East Queensland water grid, the recent privatisation of electricity
delivery in South East Queensland, or defence procurement.
In attempts to resolve the water crisis, a large number of discrete projects are underway which are
designed to work together in order to collectively improve the supply of water to Brisbane, one of
which is the provision of a desalination plant. This case study explores the use of multiple
contracts, and multiple assets to address a specific problem. A second case study is the recent
contracting of electricity supply in South East Queensland, where government retained the
generation of electricity, but contracted the supply of electricity to private firms in order to fund the
ongoing water crisis. Another pertinent case study is the procurement of the new frigate for the
Australian navy, which involves a single, large, long term contract, and the investment of significant
amounts of capital into the development of economic and social infrastructure around the naval
base in order to promote industry development (Furneaux and Brown 2007b). The different type of
actors involved in each case study and the role they take, is outlined below:
Table 3 - Types of actors involved in potential case studies
Water (in SE Queensland)
Electricity (in SE
Queensland)

Defence

Production

Public

Private

Private

Private
Public

Supply
Deployment

Both private and public

actors
Public

(applies to defence
only)

Sampling within each of the case studies would identify all of the key agents in the CAS. Rulebased action and matching of appropriate action to recognised situations on the other hand, is
seen as the province of experienced decision makers experts, who have developed capability in
a particular area over time (March and Simon 1993, 11). Snowball sampling of key informants is
seen as particularly important for research involving policymakers (Farquharson 2005). The
snowball sampling approach is useful for exploring sets of relationships, as agents in the system

are aware of each other, and thus will not exclude any important agents (Milward and Provan
1998). Agents will be identified from existing information and asked to nominate additional
organisations not on the original list, thus following a snowball methodology. An example of this
would be in the prequalification supplier arrangements which determine who is eligible to contract
for the provision of particular engineering assets.
The sample size of interviews in qualitative research is determined by theoretical saturation once
there is no new data coming forward in interview or in focus groups, then there is a probability that
theoretical saturation has been reached (Bryman and Bell 2001: 372).

Unit of analysis
Typically in organisational studies the unit of analysis focuses at a specific level, as an organisation
is viewed as a level above the individuals that form it; and the organisation is in turn below
economic and governmental forces which are seen as above and beyond the organisation and the
individuals which comprise it (Stacey and Griffin 2005). For systems analysis such discrete
analysis is problematic as processes within an organization shape the external world, even as it is
being shaped by that world (March and Simon 1993, 17). A hallmark of systems research is the
exploration of multiple agents and levels of action within a system (Holland 1995). Hitchins (2003,
80) argues that all systems are in fact comprised of subsystems so every system is a system of
systems. Holland states this another way by noting there are multiple levels within systems
(Holland 1998, 9) and higher order systems are built up from a combination of lower level systems
(Holland 1995). This concept of emergence was discussed in detail above. Given that all of the
different elements identified above appear to interact with each other, and influence each other,
then not only does this fit with the description of what a system is and how it operates, but also
helps to solve the unit of analysis question.
Systems dynamic models typically seek to examine the output of the system as a whole, rather
than the individual actions of individual agents (Robertson 2005). Lewin and Regine (2003, 169)
argue instead that the correct unit of analysis for systems research is the interactions or
relationships between people and between organisations. Hitchins (2003, 34) notes that
government systems are both frameworks of rules created by humans to regulate human
behaviour; as well as groups of people who design and implement change and are changed by the
systems they have helped to create. Thus the study of systems must study the parts, as well as the
interactions between the parts (Holland 1998, 13). Holland (1995) suggests that at least two tiers of
analysis are needed in order to understand systems of any sort one which models the set of
relationships and the flow of resources between agents at one level, and the other a higher level
which sets out the rules for interaction and adaptation of the system itself.
Consequently the unit of analysis is the procurement system the range of agents, the external
environment, the rules by which decisions are made all of the elements of the system which were
noted in the literature review. It is contended that these factors the environment, the interaction
rules and rules about the system, together with the roles of individual agents and emergent
structures are all important to understanding the system, and to explore the set of relationships, as
well as the rules and the adaptation involved in the CAS.
The following section outlines research tactics in the areas of systems thinking and system
dynamics research, which it is proposed will enable effective answering of the research questions
and is particularly important for the identification of the various agents in a system, how they
interact and how they adapt and change. Secondly the sections that follow provide an overview of
how to develop models of systems through focus groups, which is held to greatly improve the
validity of research into systems, and therefore enhances the understanding of system dynamics.

Systems dynamics research

Anderson (1999) argues that complexity thinking provides a significant and powerful new
perspective for researchers to examine decision making within organisations. As noted above,
CAS research has been advocated as a fruitful framework for examining decision making in

engineering asset procurement. Researchers who examine complex systems typically attempt to
model the dynamics of the system under investigation (Pidd 1996). A key reason for this is that the
process of model building examines and helps to explain the complexity of the system which is
being examined (Holland 1998, 4). Such a model would depict the behaviour of the system
(Hitchins 2003), and decision-making within the system (Anderson 1999). This raises the question
as to what is a model, as understood from within the complexity perspective of management. A
succinct definition of a model which shall be used in this research:
A model is a representation of reality which abstracts the features of the
situation relevant to the question being studied. The means of
representation may vary from a set of mathematical equations or a
computer program to a purely verbal description of the situation, in which
judgement alone is used to assess the consequences of various choices.
(Quade and Boucher 1968, cited in Hoos 1981, 40).
As with most business research, models of complex systems can occur in a variety of modes and
methods, and can be divided into quantitative and qualitative approaches (Cohen 1999). Each has
their own strengths and weaknesses, and applicability to different situations and contexts. An
overview of qualitative and quantitative approaches to modelling CAS and their relevance to the
examination of the procurement of engineering assets follows.
Quantitative and qualitative approaches to modelling the dynamics of CAS
Various authors have approached the computer simulation of policy outcomes from a quantitative
perspective (e.g. Mass 1991). One reason given for this is that qualitative models can become too
ambiguous and difficult to simulate manually (Sterman 1994). Against this, Hoos (1981) is strongly
critical of social scientists developing computer models as though these are definitive
representations of real world problems, and suggests that inevitably important factors under
investigation will be neglected, thereby reducing the validity of the model of the system being
developed. Emery (1981) also argues that mathematical models of complex systems ignore the
vast complexity of human action and choices.
Wolstenholme (1992) argues that having completed a qualitative system dynamics model, there is
no pressing need to convert the model to a computer based application as the diagram itself, if
done properly, can promote thinking in its own right. Richardson (1999) has argued that the
general trend in systems dynamic research is towards qualitative mapping approaches by
themselves, without quantitative simulation. Coyle (2000) agrees arguing that a rigorous qualitative
description of a system might be of significant value in and of its own right as it would lead to a
better understanding of the problem.
Given that research into nascent or little understood phenomena should utilise a qualitative,
explorative approach to the research problem (Edmondson and McManus 2007), this project will
use a qualitative approach to model the dynamics of engineering asset procurement as a CAS.
Qualitative approaches to the modelling of systems, however, come in various sub-types, with Soft
Systems Methodology (Checkland 1981, Coyle and Alexander 1997), system dynamics (Forrester
2007) and complexity theory (Klijn and Teisman 2007) all prevalent in studies of organisational
systems (Jackson 2003).
Soft systems methodology (SSM) is held to be an effective tool which can be used to model a
system (Bergvall-Kreborn, Mirijamdotter, and Basden (2004). However, Stacy and Griffin (2005)
argue that SSM is intractably linked to participant observation and action learning methodologies
which seek to change the system in which the research is a part. Given that the research problem
being explored is focussed on exploratory research, then qualitative approaches which explicitly
set out to change a system from the outset, such as SSM, would not be appropriate. Instead
research approaches which explore complexity and the dynamics of adaptive systems will be used.
Complexity and systems dynamics approaches

Systems dynamics approaches do not set out to attempt to change the system which is being
modelled. However, the very process of undertaking the research will generate causal loop
diagrams (discussed below) to participants. Mass (1986) argues that if these models of the system
are sufficiently accurate, it may move decision makers towards a deeper conceptualisation of the
impact of their policy choices, and thereby lead to a change in their choices. This feedback
process is a process termed double loop learning, which is outlined in Figure 1 below. Figure 1 is
in fact an example of a causal loop diagram which is a central element in complexity thinking and
system dynamic models.
Figure 4 - Double loop learning and decisions (Sterman 2000, 19)

Causal loop diagrams

Causal loop diagrams were used to depict the basic causal mechanisms that were seen to underlie
behaviour of agents in the system (Randers 1973, 1980). Hitchins (2003) argues that causal loop
diagrams help to depict complex problems particularly the elements, and behaviour of systems.
Causal mapping is an important way of depicting the relationships and influences between parts of
a system (Brock, Chesbro, Cragan and Klumpp 1973), and both Hitchins (2003) and Sterman
(2000) argue that causal loop diagrams are important tools for demonstrating feedback in systems.
Miles and Huberman agree, stating:
A causal network is a display of the most important independent and dependant
variables in a field of study and of the relationships among them (1994, 153)
Causal maps can be built from the ground up, or commence from an initial set of assumptions, and
can be amended following information provided by informants, with the latter generating results at a
faster rate (Miles and Huberman 1994). Causal maps can be used for both within case analysis and
between case analysis (Miles and Huberman 1994), which is important for this study, as it explores
multiple cases of engineering asset management and maintenance procurement systems. An
example of a causal loop diagram was provided in the literature review in order to explicate the
issue of co-evolution (Figure
Wolstenholme (1999) argues that qualitative modelling of systems dynamics through causal loop
diagrams enables participants to externalise mental models of the system in question. Cognitive
mapping is an important approach to understanding and depicting the mental models of decision
makers (McDonald, Daniels and Harris 2004). How these mental maps of CAS might be elicited is
discussed below.
Mental maps and models of complex systems
Farsides (2004) argues that cognitive mapping is a process which aims to develop a better
understanding of peoples worlds and how they conceptualise the various concepts which might
comprise such a world. Morecroft (2004) argues that mental models are important for developing
an understanding of how pieces in a system fit together, how the parts interact with each other.
Anderson, Meyer, Eisenhardt, Carley and Pettigrew (1999) suggest that it is important to map
agents mental models in complex systems, together with the ties that connect them together.
Edkins, Karul, Maytorena-Sanchez and Rintala. (2007) note that cognitive mapping tools are
extremely valuable in mapping complex managerial environments, suggesting that that NVIVO for
data coding and Decision Explorer for the diagrammatic representation of relationships between
concepts. Brock et al. (1973, 50) argues that mapping of mental models should include: the agents

within a system; the relationship between these agents; their goals; the rules by which they make
decisions; the outcomes of the interactions between agents in comparison to their goals.
The specific methodologies advocated for garnering this information include semi-structured
interviews, content analysis of the interviews and concept mapping (Farsides 2004; Edmondson
and McManus 2007). Vennix (1999), Luna-Reyes, Martinez-Moyano, Pardo, Cresswell, Andersen,
and Richardson (2006) and Ackermann and Eden (2004) agree but suggest that in addition to
interviews and qualitative text analysis, mapping through focus groups enables generation of more
complex maps as respondents are able to see each others models and adapt the final diagram.
This process is similar to that advocated by both Ford and Sterman (1998, 317) and Vennix and
Gubbels (1992) who recommend eliciting the mental models of multiple agents through a staged
process of firstly conducting interviews, mapping the concepts by researchers, and then checking
the models with focus groups of experts in order to further test and refine the models.
Thus eliciting experts mental maps requires the use of interviews and focus groups which fits well
with a qualitative case study approach outlined above. Such mental maps are vital to
understanding engineering asset management procurement from a CAS perspective particularly
the dynamics of the system its agents, their rules and how the system interacts.
Stated succinctly, the process for developing an understanding of CAS such as the procurement of
engineering asset management for this project will be:
o
o
o
o
o

Conduct interviews with key informants and gather relevant documents on the process
under examination
Identify key agents, goals, interactions, rules, and outcomes of interactions from these
sources
Develop initial causal loop diagrams
Conduct focus groups to check and refine the models
Conduct cross case comparisons of the cases

These steps are discussed in detail below.

Data collection instruments and rationale

In order to discover the elements and to model the CAS noted above, specific methodologies will
be used to gather data:
Policy analysis and evaluation will identify the critical elements of engineering asset
procurement, particularly the established rules which might guide the procurement process
Semi-structured interviews will be utilised to develop a fuller understanding of the range of
agents involved in the system, the roles they play, and the rules that they operate by.
Focus groups will be utilised in order to explore the mental models arising from the initial
set of interviews in order to test and refine the concept and model of procurement as a
system. Any differences between the rules available in the literature, and the rules of
individual agents will be explored and commented on.
As outlined above, such a process enables the eliciting the extent, nature, and functioning of CAS
such as engineering asset management, together with the rules used by agents to make decisions.
This project also uses multiple case studies, which Yin (2003b) argues should include multiple
sources of evidence. Multiple methods of analysis are typical in case study research and allow for
triangulation of data which is important in qualitative research to enhance validity (Eisenhardt 1989:
537). In this case the original data set will be derived from the policy documents which explore the
existing understanding of procurement as set out in practitioner and academic literature. The
interviews will generate a set of raw data, exploring roles, relationships and rules involved in the
procurement process. The primary data and the secondary data will then be examined, compared
and contrasted. An initial model of the procurement process which arises from the two data
sources will then be presented back to a focus group of nominated experts in the procurement
process. The validity of the model will be explored with the focus group in order to test for

situations and circumstances which might guide the use of a particular rule in a given situation, for
a particular asset, and not in others.
The specific methodologies of interviews and focus groups are discussed in detail below, together
with triangulation between data sources.

Semi-Structured Interviews
Semi-structured face-to-face and telephone interviews will be conducted with all agents involved in
the procurement of a particular engineering asset clients, central agencies, lead contractors, subcontractors, funding bodies, and policy activists. Interviews provide the initial set of information
about the CAS being examined, particularly the agents, their goals and their interactions with other
agents. General information about the procurement of engineering assets system would also be
generated from the interviews. Semi-structured interviewing was selected as it provides for crosscase comparability (Bryman and Bell, 2001: 346), and is important when conducting exploratory
studies particularly in order to find out what is actually happening in practice (Saunders, Lewis
and Thornhill 2000: 245). Mitzberg (1979, 587) argues that semi-structured interviews are an
important method to use in qualitative research such as is being undertaken in this project:
Semi structured interviews provide key information about policy processes, as they allow for the
telling of individual perceptions and understanding (Marinetto 1999: 72). Interviewees will be
provided with opportunities to review and correct interview summaries, by checking the data for
accuracy, thereby strengthening the internal validity of the research (Mertens 2005).
The information provided in the interviews will then enable the generation of initial model of the
CAS being examined. These initial models will be tested and further refined in a series of focus
groups.

Focus Groups
Focus groups are important in qualitative research as these allow for variance in the interpretation
of issues by participants, and to understand the ways these differences are resolved and
consensus is built (Mertens 2005). In a group interview the researcher acts as facilitator and
manager of the discussion (Saunders, Lewis and Thornhill 2000: 268). Focus group techniques
have been noted above as an important method for testing and improving the mental models of
engineering asset procurement systems developed from initial sets of interviews (Ackermann and
Eden 2004).
The sample size of focus groups would be four to six groups per case study, with 7 to 10 people
per group. These numbers are considered acceptable for answering research questions in focus
groups, although the numbers can be adjusted for specific research questions (Mertens 2005).
Approximately four to five questions can be asked effectively in a given focus group, as opportunity
for each member to participate is encouraged and discussion amongst group members will limit the
amount of topics that can be covered in a single session (Cavana, Delahaye and Sekaran 2001).
As noted above, the purpose of the focus groups is to explore the initial models of CAS developed
following the interviews, and to allow for the further development and refinement of the models.
This process is held to improve the validity of models (Miles and Huberman 1994).
Individual informants will be de-identified and any commercial in-confidence information will not
been divulged. All interviews and focus groups will be conducted in confidentiality, and the names
of interviewees will be withheld. The names of government departments, government reports, and
most government policies will not be obscured as most of this information is already freely
available, either on the Internet or in public libraries.

Triangulation
Triangulation can be used between the various data sources in order to clarify meaning, verifying
the repeatability of the observation or interpretation (Stake 2005). Triangulation can enable a

holistic understanding of the problem being investigated (Jick 1979), and minimises the risk of
potential bias that may arise if only one methodology was used (Scandura and Williams 2000,
1249). This project will undertake triangulation of data and method (Patton 1987, cited in Yin
2003b). Data triangulation uses data from different sources in order to triangulate and look for
differences. This is particularly important for model building processes. The sections above have
outlined how the use of interviews and focus groups enable triangulation between methods in order
to develop and test models of the system. Additionally, Yin (2003) argues that data can be
triangulated between cases, thereby further enhancing generalisability.

Data analysis
Huberman and Miles (1994) argue that the first step in analysis is to provide a descriptive outline of
what is happening in the case. Modelling engineering asset procurement from a CAS perspective,
entails the description of the system, its agents and their interactions, and the dynamics of how
each system operates. The extant literature and interviews will form the basis for the initial set of
causal loop diagrams. The development of these causal loop diagrams follows a pattern of
inductive reasoning which firstly observes the system, analysing patterns and themes, formulating
relationships and then develops a theory (Cavana, Delahaye and Sekaran 2001), which in this
case is a model of the system. Triangulation can be used between the various data sources in
order to clarify meaning, verifying the repeatability of the observation or interpretation (Stake
2003), which in this case is between the secondary data and the primary data. An iterative
approach is followed where data is analysed until no alternative explanation can be found (Bryman
and Bell 2001: 426).
Within case analysis is the first step in coping with a large amount of data which can be generated
in case study research (Eisenhardt 2002). Such analysis is aided through data displays which
condense key incidents or elements of the study. The provision of a dialogue between ideas and
evidence is important in the analysis phase, as it enables the opportunity of seeing relationships,
themes, patterns and clusters of data, and affords the opportunity of developing explanations about
such data.
The initial models developed from document analysis and interviews will then be discussed in a
focus group setting where the assumptions, relationships, rules and roles of each element of the
system are examined and modified. In this study, this would require the structure and dynamics of
each CAS to be elaborated, tested, refined and improved. The emerging constructs from the case
studies of each system, are tested in relation to each case and whether they hold or are different
for each case thereby strengthening the internal validity of the research (Eisenhardt 2002). The
important process is to test for the validity of the model in the minds of the key informants who
operate within the system on a daily basis.
Once each case study has been finalised, cross case comparisons are possible, which strengthen
the external validity of the research (Huberman and Miles (1994), and facilitate better understand
of CAS (Holland 1995). During this phase of the research, there is a search for patterns of
similarity or difference between cases, together with an iteration between data and emergent
theory which is strengthened and built from multiple cases and multiple data sources (Eisenhardt
2002). This process has been depicted by (Yin 2003b: 50).

Figure 5 Process for implementing a case study (Yinn 2003b: 50)

Define and design

Analyse and
conclude

Prepare, collect and analyse

Conduct
1st case
study

Write
individual
case
report

Select
cases

Modify
theory
Conduct
2nd case
study

Develop
theory

Draw crosscase
conclusions

Write
individual
case
report

Design
data
collection
protocol

Develop
Policy
implications
Conduct
any
remaining
case studies

Write
individual
case
reports

Write crosscase report

As the application of CAS theory is novel, the applicability of the theory to specific instances of
engineering asset procurement will enable the testing of the theory in multiple contexts. By testing
CAS theory in multiple cases within a single research project, cross case comparisons can be
undertaken which enable the testing and building of theory (Daneke 2005).

Summary of data collection stages

A summary of how these methodologies could be utilised in the research project is outlined below:
Table 4 - Summary of data collection strategies
Literature
Interviews
review /
Qualitative text
analysis
Identify factors
Initial interviews to
Phase 1
in literature
elicit rules, roles and
Identification of
Gather policy
relationships of
elements, agents,
documents
agents in the system
rules, roles and
tension
Phase 2
Development of
theoretical model
Phase 3 Validation
/ improvement of
the model

Recheck the
literature for
any missing
elements

Interviews on
specific case studies
to clarify issues.

Phase 4 cross
case comparisons

If needed

If needed

Focus
groups

Comments

Collate data for the

development of the initial
models.

Focus
groups to
clarify and
improve the
models
If needed

Initial analysis, synthesis

and comparison of data.
Development of initial
model.
Testing and modification
of the models

Cross case comparison of

the data, looking for
similarities and differences
between the cases in
order to build theory

Data depiction
Displaying of data and analysis of data are often relates, as robust explanation and depiction is
likely to lead to better analysis. The literature on researching systems provides useful insights into
the display and depiction of the agents in a CAS, and how they interrelate.

Matrices and charts

Components of the system need to be identified and explained importantly that there may be subcomponents which act relatively independently of each other, and yet have an impact on the
outcome as a whole (Brock et al. 1973, 62). N2 charts are forms of matrixes which are useful for
depicting complex sets of relationships and interaction between different components of a system,
such as the sharing of resources (information or financial) (Hitchins 2003) (see Figure 3 below).
Figure 6 An example of a N2 matrix (adapted from Hitchins 2003, 145)

Sub-system A
0

0
Sub-system B

0
Sub-system C
0

0
Sub-system D
0
0

0
0
Sub-system E
0

0
0
0
Sub-system F

Gill (1996) argues that by firstly eliciting relationships from interviews and placing these onto a
matrix form of diagram, relationships between parties can be identified. These can then be turned
into system dynamics diagrams, causal maps, or other forms of network diagram. Richardson
(2005) notes that from these matrixes it is possible to build network graphs of relationships
between agents in a network, which will enable the understanding of the set of agents in a CAS
and how they interrelate to each other.
Developing of models through iteration
As noted above, once the initial models have been developed, it is important to gain feedback from
informants on the model under development in multiple times (Miles and Huberman 1994). The
development of initial diagrams is an important first step in understanding and depicting the
interaction and dynamics of the specific instance of engineering asset procurement from a CAS
perspective. Wolstenholme (1992, 129) argues that the development of models which analyse
system dynamics, should involve several iterations of the model. This process will be undertaken
through the focus groups noted above. Homer (1996) argues that iteration is important in order to
ensure the validity of the model, and that researchers should not be surprised if there are a number
of stages to the development of a model process.
Homer and Oliva (2001) argue that the model of a CAS needs to be revisited on numerous
occasions, in order to ensure that the model replicates reality. As Sterman (1994) argues without
modelling we might think that we are learning to think holistically, when we are actually learning to
jump to conclusions. Richardson (1996) argues that this process of checking the model with key
informants of experts ensures the validity of the mental models.
Data depiction and iteration of findings from informants are thus an important part of the process of
developing a qualitative model of the engineering asset management procurement from a CAS
perspective.

Validity, reliability and generalisability

Validity, reliability, and generalisability are hallmarks of good research, however, these aspects can
be difficult to implement in qualitative studies. Yin (2003b) outlines how validity and reliability can
be enhanced in case study designs:
o Construct validity multiple sources of evidence, have key informants review drafts
o Internal validity pattern matching and explanation building, address rival explanations
o External validity use replication logic in multiple case studies
o Reliability use case study protocol, case study database
o Transferability parallels external validity in quantitative studies, and relies upon multiple
cases in the one study.

As noted above, multiple sources of evidence will be used to improve the validity of constructs, and
key informants will also be allowed to review summaries of interviews (Yin 2003b). By matching
patterns across interviews and addressing rival explanations in the focus groups, internal validity
will be strengthened (Yin 2003b). As multiple case studies will be used, replication of the same
methodology and logic can improve the external validity, reliability and transferability (Yin 2003b).
Thus the methodology, outlined above, by using multiple case studies, multiple sources of
evidence and involving informants in checking interviews and modes, will improve the validity,
reliability and generalisability of the findings (Yin 2003a).
While it may not be possible to generalise from four case studies to every other case of
engineering asset procurement due to the specific nature of each system (Klein and Teisman
2007), it should be possible to generalise from the case studies back to theory (Yin 2003a). The
Cooperative Research Centre for Integrated Engineering Asset Management (CIEAM) will provide
a highly useful dissemination role by virtue of access to different industry partners which comprise
the Centre

Limitations
Some data is likely to be sensitive and classified depending on the nature of the research, or
may not be able to be made public. This may possibly affect the selection, although certainly the
reporting, of a specific case study. However, industry partners associated with CIEAM have
indicated that they would be interested in participating in the research, with some indicating that
they would facilitate access to their entire supply chain. As noted in the sampling section above,
the research will select across multiple asset classes in order to generate a better understanding of
the various approaches to engineering asset procurement.

Assumptions
Industry partners with the CRC for Integrated Engineering Asset Management (CIEAM) have
indicated interest in the project. The Department of Defence and the agency responsible for the
Collins class submarines (ASC) have both expressed interest in participating. The Department of
Defence is exploring CAS framework in a range of areas, including procurement. CIEAM has
provided financial support to this project and will broker access to these research partners in order
to facilitate access to research partners.
Cooperation by industry and government is necessary in order to gain access to interviewees and
focus group participants. If cooperation is not possible for various reasons then alternative case
studies from other members of CIEAM will be substituted in place of those above.

Reference List
Ackermann, F. and C. Eden. 2004. Using causal mapping individual and group, traditional and new In Systems
Modelling: Theory and practice, ed. M. Pidd, 127145. Chichester: John Wiley and Sons.
Adriani, P. (2003) Evolutionary dynamics of industrial clusters. In Mitleton-Kelly, E. (Ed) Complex systems and
evolutionary perspectives on organisations: The application of complexity theory to organisations. Amsterdam:
Pergamon, Elsevier, pp. 127146.
Albino, V., Carbonara, N., & Giannoccaro, I. (2005). Industrial districts as Complex adaptive Systems: Agent Based
Mdoels of Emergent Phenomena. In C. Karlsson, B. Johansson & R. R. Stough (Eds.), Industrial Clusters and
Inter-Firm Networks (pp. 58-82). Cehltenham, UK: Edward Elgar.
Anderson, P. 1999. Complexity theory and organizational science Organizational Science 10(3): 216232.
Anderson, P., A. Meyer, K. Eisenhardt, K. Carley and A. Pettigrew. 1999. Introduction to the Special Issue: Application of
complexity theory to organizational science Organizational Science 10(3): 233236.
Anderson, R.A., Crabtree, B.F., Steele, D.J., McDaniel, R.R. (2005) Case study research: The view from complexity
science Qualitative Health Research 15(5), 669-685.
APCC see Australian Procurement and Construction Council
Australian Procurement and Construction Council. 2001. Asset Management. Canberra: APCC. accessed from
http://www.apcc.gov.au/docs/AssetManagement2001%5B1%5D.pdf on 22 December 2007.
Australian Procurement and Construction Council. 2002. Client Skills: Skills required by Government as the Construction
Industry Client. Canberra: APCC. Downloaded from http://www.apcc.gov.au/docs/ClientSkillsSep2002.pdf on
10 February 2006.
Australian Procurement and Construction Council. 2003. National Procurement Reform Principles. Canberra: APCC.
Accessed from http://www.apcc.gov.au/docs/APCCNatProcurementReformPrinciplesMch2003.pdf on 22
December 2007.
Australian Procurement and Construction Council. 2007. Australian and New Zealand Government Procurement
Framework for Sustainable Procurement. Canberra: APCC.
http://www.apcc.gov.au/publications.asp?pid=156#jump-156 on 22 December 2007.
APCC (2008) Australian Procurement and Construction Council, available at: http://www.apcc.gov.au (accessed January
2008).
Association of Local Government Engineering New Zealand Inc and Institute of Public Works Engineering of Australia.
2002. International Infrastructure Management Manual. Thames: INGENIUM.
Austen, S., R. Seymour, K.A. Brown, C.W. Furneaux, and A. McCabe. 2007. Multi-Outcomes Construction Policies:
Final Report (First Draft). Brisbane: Icon.Net Let. [unpublished industry report]
Australian Public Service Commission. 2003. The Australian Experience of Public Sector Reform: Chapter 7 Structural
Reform. Downloaded from http://www.apsc.gov.au/about/exppsreform9.htm, on 18 May 2005.
Babbie, E. 2004. The practice of social research. 10th edition. Belmont, CA: Wadsworth/Thompson Learning.
Ball, P. 2005. Critical mass: how one thing leads to another London: Arrow.
Barlow, J. and M. Rber. 1996. Steering not rowing: Coordination and control in the management of public services in
Britain and Germany International Journal of Public Sector Management 9(5/6): 73 89.
Bar-Yam, Y. 2006. Improving the effectiveness of health care and public health: A multiscale complex systems analysis.
American Journal of Public Health Vol. 96(3): 459466.
Bergvall-Kreborn, B. A. Mirjamdotter and A. Basden. 2004. Basic principles of SSM modelling: An examination of
CATWOE from a soft perspective. Systemic practice and action research, 17(2): 5573.
Blackman, T. 2001. Complexity theory and the new public management Social Issues 1(2) www.whib.co.uk/socialissues.
Bovaird, T. 2006. Developing new forms of partnership with the market in the procurement of public services. Public
Administration Vol. 84(1): 81102.
Bovaird, T. 2007. Exploring the governance mechanisms in complex adaptive systems: The case of UK local
government, Panel Track - Complexity Theory and Public Management International Research Symposium for
Public Management April, 2-4, 2007,
http://www.irspm2007.org/Fullpaper/PT_Complexity%20Theory_Bovaird.pdf downloaded on 15 November
2007.
Bridgman, P. and G. Davis. 2004. The Australian Policy Handbook. Crows Nest, NSW: Allen and Unwin.
British Standards 2003. PAS 55-1 Asset management: Specification for the optimised management of physical
infrastructure assets. London: UK.

Brock, B.L. J.W. Chesbro, J.F. Cragan and J.F. Klumpp. 1973. Public Policy Decision Making: Systems analysis and
comparative advantages debate. New York: Harper and Row
Brown, K.A., C.W. Furneaux, S. Janssen and D. Allan. 2008. Harmonisation of OH&S Regulation in Australia: An
Evaluation of three initiatives Third International Conference of the Cooperative Research Centre (CRC) for
Construction Innovation Clients Driving Innovation: Benefiting from Innovation (12-14 March 2008) [full paper
accepted without revisions].
Bryman, A. and E. Bell. 2003. Business Research Methods. Oxford: Oxford University Press.
Cavana, R. and L.C. Clifford. 2006. Demonstrating the utility of system dynamics for public policy analysis in New
Zealand: the case of excise tax policy on tobacco. System Dynamics Review 22(4): 321 348.
Cavana, R.Y., B.L. Delahaye and U. Sekaran. 2001. Applied Business Research: Qualitative and Quantitative Research
Methods. Milton, QLD: John Wiley and Sons.
Integrated Engineering Asset Management (2006) CIEAM Profile Volume Two 2006 downloaded from
http://www.cieam.com/documents/cieam_profile_news_vol2_06.pdf on 16 March 2008.
Chalmers, J. and G. Davis. 2001. Rediscovering Implementation: Public sector contracting and human services
Australian Journal of Public Administration, Vol 60(2): 7485.
Chapman, J. 2002. System failure: Why governments must learn to think differently. London: Demos. Downloaded

from www.demos.co.uk on 11 December 2007.

Chapman, R. 1989. Federalism, inter-governmental relations and the Australian policy process. In Governing
Federations: Constitution, Politics, Resources, eds. M. Wood, C. Williams and C. Sharman. Sydney: Hale and
Iremonger Pty Ltd.
Checkland, P.B. 1981. Systems thinking, Systems practice. Chichester: Wiley.
Chu, D., R. Strand and R.P. Jelland. 2003. Theories of complexity: Common denominators if complex systems.
Complexity 8(3): 1930.
CIEAM see Centre for Integrated Engineering Asset Management
Clippinger, J.H. III (1999) Order from the bottom up: Complex adaptive systems and their management. In J.H.
Clippinger (ed) The Biology of Business San Francisco: Jossey Bass, 1-25.
Coghill, K. 2004. Federalism: Fuzzy Global Trends. Australian Journal of Politics and History, 50(4): 4156.
Cohen, M. 1999. Commentary on the Organization Science Special Issue on Complexity, Organization Science, 10(3):
373376.
Coyle, G. 2000. Qualitative and quantitative modelling in system dynamics: some research questions, System Dynamics
Review, 16(3): 225244.
Coyle, G. 2001. Rejoinder to Homer and Oliva, System Dynamics Review, 17(4): 357363.
Coyle, R.G. and M.D.W. Alexander. 1997. Two approaches to qualitative modelling of a nations drug trade, System
Dynamics Review, 13(3): 205222.
Daneke, G.A. 2001. Sustainable development as systemic choices, Policy Studies Journal, 29(3): 514532.
Daneke, G.A. 2005. The Reluctant Resurrection: New Complexity Methods and Old Systems Theories, International
Journal of Public Administration, 28(1): 89106.
Demirag, I. 2004. Towards Better Governance and Accountability: Exploring the Relationships between the Public,
Private and the Community, Journal of Corporate Citizenship, 15: 19 26.
Denzin, N.K. and Y.S. Lincoln. (Eds.) 2003. Strategies of Qualitative Enquiry. 2nd Edition. Thousand Oaks: Sage.
Department of Treasury and Finance. 2005. Strategic Asset Management Framework
http://www.dtf.wa.gov.au/cms/uploadedFiles/01_samf_samf_082005.pdf accessed on 23 December 2007.
Domberger, S. 1994. Public Sector Contracting: Does it Work? The Australian Economic Review, 3rd Quarter: 91-96.
Doyle, J.K. and D.N. Ford. 1998. Mental models concepts for system dynamics research, System Dynamics Review,
14(1): 329.
Edkins, A., E. Kurul, E. Maytorena-Sanchez and K. Rintala. 2007. The application of cognitive mapping methodologies in
project management research, Project Management, 25: 762772.
Edmonson, A.C. and S.E. McManus. 2007. Methodological fit in management field research, Academy of Management
Review, 43(4): 11551179.
Eisenhardt, K.M. 1989. Agency Theory: As Assessment and Review, Academy of Management Review, 14(1): 57 74.
Eisenhardt, K.M. 2002. Building theories form case study research. In The Qualitative Researchers Companion, eds.
A.M. Huberman and M.B. Miles, 531. Thousand Oaks: Sage Publications.
Eisner, M.A., J. Worsham and E. Ringquist. 1996. Crossing the organizational void: The limits of agency theory in the
analysis of regulatory control, Governance: An International Journal of Policy Administration, 9(4): 407 428.
Emery, F.E. 1981. Adaptive systems for our future governance. In Systems Thinking Volume Two, ed. F.E. Emery, 391
408. Harmondsworth, UK: Penguin Books.

Farquharson, K. 2005. A different kind of snowball: Identifying key policymakers, International Journal of Social
Research Methodology, 8(4): 345 353.
Farsides, T. 2004. Cognitive mapping: Generating theories of psychological phenomena from verbal accounts and
presenting them diagrammatically. In Doing social psychology research, ed. G.M. Breakwell, 255288. Oxford:
British Psychological Society and Blackwell Publishing.
Ferguson, A. 2007. $200bn infrastructure bonanza coming The Weekend Australian December 15-16, 2007: 36.
Ford, D.N. and J.D. Sterman. 1998. Expert knowledge elicitation to improve formal and mental models, System
Dynamics Review, 14(4): 309340.
Forrester, J.W. 2007. System dynamics the next fifty years, System Dynamics Review, 23(2/3): 359370.
Furneaux, C.W. and K.A. Brown. 2007a. Capabilities, institutions and markets: A cross jurisdictional analysis of
embedded public values in public works procurement in Australia. Infrastructure Policies and Public Value panel
track, International Research Symposium on Public Management XI 2nd to 4th April 2007 Potsdam
(Germany). (accepted for publication in Journal of Public Management and Money)
Furneaux, C.W. and K.A. Brown. 2007b. Final Report - Stage Two of the Australian Government Building and
Construction OHS Accreditation Scheme. Unpublished report prepared for the Office of the Federal Safety
Commissioner. [report accepted]
Furneaux, C.W., K.A. Brown, D. Allan, S. McConville, S. McFallan, K. London and J. Burgess. 2006b. Client capabilities
and capital works procurement policies: A comparative analysis of Australian jurisdictions. In Clients Driving
Construction Innovation: Moving Ideas into Practice, eds K.A. Brown, K. Hampson and P. Brandon, 6271.
Brisbane: Cooperative Research Centre for Construction Innovation, Icon.Net Pty Ltd.
Furneaux, C.W., K.A. Brown, D. Allan, N. Abel, S. McConville, S. McFallan, K. London and J. Burgess. 2006a. Capital
works procurement policies in Australia: implications for understanding and meeting client needs. In
Proceedings Clients Driving Innovation: Moving Ideas into Practice, Gold Coast, Australia.
Gill, R. 1996. An integrated social fabric matrix/system dynamics approach to policy analysis System Dynamics Review
,12(3): 167181.
Globerman, S. and A. Vining. 1996. A Framework for Evaluating the Government Contracting-Out Decision with an
Application to Information Technology, Public Administration Review, 56(6): 557-586.
Griffin, D. and P. Shaw. 2000. Complexity and Management: Fad or Radical challenge to systems thinking. London:
Routledge.
Haynes, P. 2007. Complexity theory and evaluation in public management: a qualitative systems approach. Panel Track
- Complexity Theory and Public Management, International Research Symposium for Public Management April,
2-4, 2007. http://www.irspm2007.org/Fullpaper/PT_Complexity%20Theory_Haynes.pdf downloaded 15
November 2007.
Hellstrm, T. 2007. Critical infrastructure and systemic vulnerability: Towards a planning framework, Safety science, 45:
415430.
Herder, P.M. and Z. Verwater-Lukszo. 2006. Towards next generation infrastructures: an introduction to the contributions
in this issue. International Journal of Critical Infrastructures 2(2/3): 113120.
Hitchins, D.K. 2003. Advanced systems thinking, engineering and management. Boston: Artch House.
Holland, J.H. 1995. Hidden Order: How adaptation builds complexity. New York: Basic Books
Holland, J.H. 1998. Emergence: From chaos to order. Oxford: Oxford University Press.
Homer, J. and R. Oliva. 2001. Maps and models in systems dynamics: A response to Coyle, System Dynamics Review,
17(4): 347355.
Homer, J.B. 1996. Why we iterate: scientific modelling in theory and practice, System Dynamics Review ,12(1):119.
Hood, C. 1991. A public management for all seasons? Public Administration, 69: 319.
Hoos, I.R. 1981. Methodology, methods and models In Systems thinking Volume Two, ed. F.E. Emery, 40-55.
Harmondsworth, UK: Penguin Books.
Huberman, A.M. and M.B. Miles. 1994. Data management and analysis methods. In Handbook of Qualitative Research,
eds. N.K. Denzin and Y.S. Lincoln, 428443. Thousand Oaks: Sage Publications.
INGENIUM see Association of Local Government Engineering New Zealand Inc.
Introna, L.D. 2003. Complexity theory and organisational intervention? In Complex systems and evolutionary
perspectives on organisations: The application of complexity theory to organisations, ed. E. Mitleton-Kelly, 205
219. Amsterdam: Pergamon, Elsevier.
Jackson, M.C. 2003. Systems Thinking: Creative Holism for Managers. Chichester: John Wiley and Sons.
Jensen, M.C. and W.H.J. Meckling. 1994. The Nature of Man, Journal of Applied Corporate Finance, 7(2): 419.
Jensen, M.C. 1994. Self-Interest, Altruism, Incentives and Agency Theory, Journal of Applied Corporate Finance, 7(2):
40 45.

Jensen, P. and R. Stonecash. 2004. The Efficiency of Public Sector Outsourcing Contracts: A Literature Review,
Melbourne Institute Working Paper No. 29/04, Melbourne Institute of Applied Economic and Social Research,
University of Melbourne.
Jick, T.D. 1979. Mixing qualitative and quantitative methods: Triangulation in action, Administrative Science Quarterly,
24: 602 611.
Keast, R. K.A. Brown and M. Mandell. 2007. Getting the right mix: Unpacking integration meanings and strategies,
International Public Management Journal, 10(1): 9-33.
Keast, R. M. Mandell and K.A. Brown. 2006. Mixing state, market and network governance modes: The role of
government in crowded policy domains, International Journal of Organizational Theory and Behavior, 9(1): 2750.
Kooiman , J. 1999 . Social-Political Governance: Overview, Reflections and Design, Public Management, 1(1): 6792.
Klijn, E. (2001) Rules as institutional context for decision making in networks: The approach to postwar housing districts
in two cities. Administration and Society 33(3), 133-164.
Klijn, E. (2007) Managing complexity: Achieving the impossible? Management between complexity and stability: a
network perspective Critical Policy Analysis 1(3), 252-277.
Klijn, E-H. and G. Teisman. 2007. Complexity theory and Public Administration: A critical appraisal Panel Track Complexity Theory and Public Management International Research Symposium for Public Management April,
2-4, 2007, http://www.irspm2007.org/Fullpaper/PT_Complexity%20Theory_Klijn%20Teisman.pdf downloaded
on 15 November 2007.
Klijn, E-H. and J.F.M. Koppenjan. 2006. Institutional design: Changing institutional features of networks, Public
Management Review, 6(1): 141160.
Koppenjan, J.F.M. 2001. Project development in complex environments: Assessing safety in design and decision
making, Project Development in Complex Environments, 9(3): 121130.
Ldre, O., K. Austeng, T.I. Haugen and O.J. Klakegg. 2006. Procurement routes in public building and construction
projects, Journal of Construction Engineering and Management, July 2006: 689696.
Lewin, R and B. Regine. 2003. The core of adaptive organisations. In Complex systems and evolutionary perspectives
on organisations: The application of complexity theory to organisations, ed. E. Mitleton-Kelly, 167184.
Amsterdam: Pergamon, Elsevier.
Luna-Reyes, L. I.J. Martinez-Moyano, T.A. Pardo, A.M. Cresswell, D.F. Andersen and G.P. Richardson. 2006. Anatomy
of a group model-building intervention: building dynamic theory from case study research, System Dynamics
Review, 22(4): 291320.
Main Roads Queensland. 2006. Roads Implementation Program 2006-07 to 2010-11
http://www.mainroads.qld.gov.au/MRWEB/PROD/Content.nsf/DOCINDEX/Roads+Implementation+Program+2
006-07+to+2010-11?OpenDocument, accessed on 28 April 2007
March, J.G. and J.P. Olsen. 1984. The New Institutionalism: Organisational Factors in Political Life, The American
Political Science Review, 78(3): 734749.
March, J.G. and H.A. Simon. 1993. Organizations 2nd edition. Norwich, UK: Blackwell Business.
Marinetto, M. 1999. Studies of the policy process: A case analysis. London: Prentice Hall Europe.
Markose, S.M. 2005. Computability and evolutionary complexity: Markets as complex adaptive systems (CAS), The
Economic Journal ,115: F159F192.
Mass, N.J. 1986. Methods of conceptualisation, System Dynamics Review, 2(1): 7680.
Mass, N.J. 1991. Diagnosing surprise model behavior: a tool for evolving behavioral and policy insights, System
Dynamics Review, 7(1): 6886.
McDonald, S., K. Daniels and C. Harris. 2004. Cognitive mapping in organizational research In Essential Guide to
Qualitative Methods in Organizational Research, eds C. Cassell and G. Symon, 7385. London: Sage
Publications.
McIntyre, J. and M. Pradham. 2003. A systemic approach to addressing the complexity of energy problems, Systemic
Practice and Action Research, 15(3): 213223.
Meek, J., J. De Ladurantey and W.H. Newell. 2007. Complex systems, governance and policy administration
consequences, Emergence: Complexity and Organisation, 9(1): 2436.
Mertens, D.M. 2005. Research and Evaluation in Education and Psychology: Integrating diversity with Quantitative,
Qualitative and mixed methods. 2nd Edition. Thousand Oaks: Sage Publications.
Midgley, G. and K.A. Richardson. 2007. Systems thinking for community involvement in policy analysis, Emergence:
Complexity and Organization, 9(1/2): 167183.
Milward, H.B. and K.G. Provan. 1998. Measuring network structure, Public Administration, 76: 387407.
Mintzberg, H. 1979. An emerging strategy of direct research, Administrative Science Quarterly, 24: 580589.

Mitchell, R.K., B.R. Agle and D.J. Wood. 1997. Toward a theory of stakeholder identification and salience: Defining the
principle of who and what really counts, Academy of Management Review, 22(4): 853 886.
Mitleton-Kelly, E. 2003. Ten principles of complexity and enabling infrastructures In Complex systems and evolutionary
perspectives on organisations: The application of complexity theory to organisations, ed E. Mitleton-Kelly, 23
50. Amsterdam: Pergamon, Elsevier.
Morecroft, J. 2004. Mental models and learning in system dynamics practice In Systems Modelling: Theory and practice,
ed. M. Pidd, 101126. Chichester: John Wiley and Sons.
Morgan, G. 1997. Images of Organization Thousand Oaks: SAGE Publications.
Moxnes, E. 2000. Not only the tragedy of the commons: misperceptions of feedback and policies for sustainable
development, System Dynamics Review, 16(4): 325348.
Namatame, A. and Sasaki, T. (1998) self-organisation of complex adaptive systems as a society of rational agents
Artificial Life Robotics 2, 189-195.
New South Wales Government 2005. Procurement Methodology Guidelines for Construction
http://www.managingprocurement.commerce.nsw.gov.au/nsw_government_guidelines/procurement_methodolo
gy.pdf accessed on 23 December 2007.
Niksanen, W.A. 1998. Policy analysis and public choice: Selected papers by William A Niskanen. Edward Algar:
Cheltenham.
Osborn, R.N. and J. Hagedoorn. 1997. The institutionalization and evolutionary dynamics of interorganizational alliances
and networks, Academy of Management Journal, 40(2): 261-279.
Osborne, D.E. and T. Gaebler. 1992. Reinventing government : How the entrepreneurial spirit is transforming the public
sector. Reading, Mass : Addison-Wesley.
Osbourne, S.P. and K.A. Brown. 2005. Managing change and innovation in public service organizations. New York :
Routledge.
Parsons, B.A. 2007. The state of methods and tools for social systems change, American Journal of Community
Psychology, 39: 405409.
Pidd, M. 1996. Tools for thinking: Modelling in Management Science. Chichester: John Wiley and Sons.
Pierce, J.C. 2000. Paradox of physicians and administrators in health care organisations, Health Care Management
Review, 25(1): 728.
Price, I. and F. Akhlaghi. 1999. New patterns in facilities management: industry best practice
and new organisational theory, Facilities, 17(5/6): 159166.
Provan, K.G. and P. Kenis. 2007. Modes of Network Governance: Structure, Management and Effectiveness, Journal of
Public Administration Research and Theory, [in press].
Queensland Treasury 2003. Asset Strategic Plan Guidelines
http://www.treasury.qld.gov.au/office/knowledge/docs/asset-planning/asset-strategic-plan-guidelines.pdf
accessed on 23 December 2007
Quiggin, J. 1996. Private sector involvement in infrastructure projects, Australian Economic Review, 96(1): 5164.
Randers, J. 1977. The Potential in Simulation of Macro-Social Processes, or How to be a Useful Builder of Simulation
Models. Gruppen for Ressursstudier, Oslo, Norway.
Randers, J. 1980. Elements of the System Dynamics Method. Cambridge, MA: MIT Press.
Rhodes, M. L., and G. MacKechnie. 2003. Understanding public service systems: Is there a role for Complex Adaptive
Systems Theory? Emergence, 5(4): 57-85.
Rhodes, M.L. 2007. Complexity and emergence in Public Management: The case of urban regeneration in Ireland. Panel
Track - Complexity Theory and Public Management International Research Symposium for Public Management
April, 2-4, 2007, http://www.irspm2007.org/Fullpaper/PT_Complexity%20Theory_Rhodes.pdf downloaded on
15 November 2007.
Rhodes, R.A.W. 1997. From marketisation to Diplomacy: Its the mix that matters, Australian Journal of Public
Administration, 56(2): 4053.
Richardson, 1999. Reflections on the future of systems dynamics, Journal of the Operational Research Society, 50: 440449.
Richardson, G.P. 1996. Problems for the future of system dynamics, System Dynamics Review 12(2): 141157.
Richardson, G.P. 1997. Problems in causal loop diagrams revisited, System Dynamics Review, 13(3): 247252.
Richardson, K. 2006. Complex systems thinking and its implications for policy analysis in Handbook of Decision Making,
ed. G. Morl, 189221. Boca Raton: Taylor and Francis Group.
Richardson, K.A. 2004a. Systems theory and complexity: Part 1, Emergence: Complexity and Organizations 6(3): 7579.
Richardson, K.A. 2004b. Systems theory and complexity: Part 2, Emergence: Complexity and Organizations 6(4): 7782.
Richardson, K.A. 2005. Systems theory and complexity: Part 3, Emergence: Complexity and Organizations 7(2): 104
114.

Richardson, K.A. 2007. Systems theory and complexity: Part 4. Emergence: Complexity and Organizations 9(1-2), p.
166.
Rinaldi, S.M., J.P, Peerenboom and T.K. Kelly 2001. Identifying, understanding, and analysing critical infrastructure
interdependencies, IEEE Control Systems Magazine. December 2001: 1125.
Robertson, D.A. 2005. Agent based models to manage the complex. In Managing Organizational Complexity:
Philosophy, Theory, Application, ed. K.A. Richardson, 419432. Greenwich, Connecticut: Information Age
Publishing.
Robson, A.J. 2005. Complex evolutionary systems and the red queen, The Economic Journal, 115: F211224.
Sanderson, I. 2000. Evaluation in complex policy systems, Evaluation 6(4): 433 454.
Saunders, C. 1994. Australian Economic Union in Economic Union in Federal Systems, eds. A. Mullins and C.
Saunders. Sydney: The Federation Press.
Saunders, M. Lewis P. and Thornhill, A. 2000. 2nd Edition. Research methods for business students. Harlow, Essex,
England: Pearson Education Limited.
Scandura, T.A. and E.A. Williams. 2000. Research Methodology in Management: Current practices, trends, and
implications for future research, Academy of Management Journal, 43(6): 1248 1264.
Schofield, J.W. 2002. Increasing the generalizability of qualitative research. In The Qualitative Researchers Companion,
eds. A.M. Huberman and M.B. Miles, 171203. Thousand Oaks: Sage publications.
Smith, Adam (1971) The Wealth of Nations. Volume 2. London: Everymans Library.
Smith, M.Y. and R. Stacey. 1997. Governance and Cooperative Networks: An adaptive systems perspective,
Technological Forecasting and Social Change, 54: 7994.
Snowden, D. and Stanbridge, P. (2004) The landscape of management: Creating the context for understanding social
complexity Emergence: Complexity and Organizations 6(1-2), pp. 140-148.
Stacey, R. and D. Griffin. 2005. Introduction: researching organizations from a complexity perspective. In A Complexity
Perspective on Researching Organizations: Taking experience seriously, eds. R. Stacey and D. Griffin, 112.
London: Routledge.
Stake, R.E. 2003. Case studies. In Strategies of Qualitative Enquiry, 2nd Edition, eds. in N.K. Denzin and Y.S. Lincoln,
134164. Thousand Oaks: Sage.
Stake, R.E. 2005. Qualitative case studies. In The Sage Handbook of Qualitative Research Methods, 3rd Edition, eds in
N.K. Denizn and Y.S. Lincoln, 443466. Thousand Oaks: Sage.
Sterman, J.D. 1994. Learning in and about complex systems, System Dynamics Review, 10: 291330.
Sterman, J.D. 2000. Business Dynamics: Systems thinking and modelling for a complex world. Boston: Irwin, McGrawHill.
Tilebein, M. (2006) A complex adaptive systems approach to efficiency and innovation Kybernetes 35(7/8), 10871099.
Van Hinte, T., T.I. Gunton and J.C. Day. 2007. Evaluation of the assessment process for major projectsL a case study of
oil and gas pipelines in Canada, Impact assessment and Project Appraisal, 25(5): 123137.
Vennix, J.A.M. 1999. Group model-building: tackling messy problems, System Dynamics Review, 15(4): 379401.
Vennix, J.A.M. and J.W. Gubbels. 1992. Knowledge elicitation in conceptual model building: A case study in modelling a
regional Dutch health care system, European Journal of Operational Research, 59: 85101.
Verspaandonk, R. 2001. Outsourcing For and Against, Current Issues Brief, 18 (26 June 2001) Downloaded from
http://www.aph.gov.au/library/pubs/cib/2000-01/01cib18.htm, on 18 May 2005.
Western Australia Department of Treasury and Finance. 2005. Strategic Asset Management Framework for Western
Australian Public Sector Agencies. Downloaded from
http://www.dtf.wa.gov.au/cms/uploadedFiles/00_samf_overview_082005.pdf on 15 November 2007.
Whetten, D.A. 1989. What constitutes a theoretical contribution? The Academy of Management Review ,14(4): 490495.
White, L. 2000. Changing the whole system in the public sector, Journal of Organisational Change Management, 13(2):
162177.
White, L. 2001. Effective Governance Through Complexity Thinking and Management Science, Systems Research and
Behavioral Science, 18: 241257.
Wollin, D. and C. Perry. 2004. Marketing management in a complex adaptive system: An initial framework, European
Journal of Marketing, 38(5/6): 556572.
Wolstenholme, E.F. 1999. Qualitative vs Quantitative modelling: The evolving balance, Journal of the Operational
Research Society, 50: 422-428.
Wolstenholme, E.F. 1992. The definition and application of a stepwise approach to model conceptualisation and
analysis, European Journal of Operational Research 59: 123-136.
Yin, R.K. 2003a. Case Study research: Design and Methods. 3rd Edition. Thousand Oaks: Sage Publications.
Yin, R.K. 2003b. Applications of Case Study Research, 2nd Edition. Thousand Oaks: Sage Publications.
Zikmund, W.G. 2003. Business Research Methods. Mason, OH: Thompson, Southwestern.